Brianary

Ice Cream Sandwich on the Nexus S Update

2012-01-24T21:43:00.001-08:00

Bluetooth eventually crashes in ICS, irredeemably. Only a full reboot will produce sound via Bluetooth after this happens. This makes ICS the ideal platform for spending five minutes trying to get three minutes' worth of content to play.
The bookmarks widget, which displays the browser bookmarks as a scrolling list of icons, its easily the best new widget, making access to frequently visited sites much faster.
Notifications rarely seem to notify!
Google Voice and Messaging (the texting app), in particular, are terrible about indicating something has happened, but notifications in general don't seem consistent about providing notification sounds or activating the screen.
No more search provider narrowing!
In previous Android versions, you could tap the icon to the left of the system search field, and choose which Searchable Items to use: just Applications or Browser or Music or Twitter or whatever, instead of all of them, which is slower and often too crowded to show the item you wanted. This has completely ruined the "command-line" aspect of search for me.
Picasa integration is still completely broken for uploads!
No albums are listed and trying to upload displays a "Failed to retreive account information" error. There are indications in some forums that this is a result of Picasa trying to use the wrong authentication account, but no adequate fix yet. This also highlights another problem: Who do you call for a fix, or where do you go to even to report a problem like this?
No unhide UI for contact providers.
When trying to delete contact entries automatically created by the Twitter app for everyone I follow, the app informs me that they will only be hidden. After realizing that I can just hide the contacts from that provider, there's no way to restore those "hidden" contacts short of deleting the app's data and starting over.
Much better browser, still no extensions though.
The stock Android browser seems much better, even though it offers no extensions, and isn't really a full Chrome implementation.
Sharing UI in the Gallery app is better.
In previous versions of Gallery, the sharing menu was difficult to use once too many apps registered as sharing targets. The new version handles long lists of sharing targets easily, and even offers a very good MRU interface.
System settings are available from the notification tray.
This makes turning WiFi off and on, among other things, much easier to do.
The Market is available in the corner of the app tray.
This makes getting to the market easier.
Widgets are available as a tab in the app tray.
This allows you to preview what widgets look like before adding them.
Scrollable widgets are handy, though most that I tried tend to crash.

In general, I'd say the experience was negative, and I'd recommend avoiding this upgrade until some of the problems are addressed.

STOP #SOPA/#PIPA

2012-01-18T08:27:00.000-08:00

Learn more, participate, donate.

Ice Cream Sandwich on Nexus S so far

2012-01-13T09:15:00.001-08:00

Notifications rarely seem to notify!
Google Voice, in particular, is terrible about indicating something has happened, but notifications in general don't seem consistent about providing notification sounds or activating the screen.
No more search provider narrowing!
In previous Android versions, you could tap the icon to the left of the system search field, and choose which Searchable Items to use: just Applications or Browser or Music or Twitter or whatever, instead of all of them, which is slower and often too crowded to show the item you wanted. This has completely ruined the "command-line" aspect of search for me.
Picasa integration is broken for uploads!
No albums are listed and trying to upload displays a "Failed to retreive account information" error. There are indications in some forums that this is a result of Picasa trying to use the wrong authentication account, but no adequate fix yet. This also highlights another problem: Who do you call for a fix, or where do you go to even to report a problem like this?
No unhide UI for contact providers.
When trying to delete contact entries automatically created by the Twitter app for everyone I follow, the app informs me that they will only be hidden. After realizing that I can just hide the contacts from that provider, there's no way to restore those "hidden" contacts short of deleting the app's data and starting over.
Much better browser, still no extensions though.
The stock Android browser seems much better, even though it offers no extensions, and isn't really a full Chrome implementation.
Sharing UI in the Gallery app is better.
In previous versions of Gallery, the sharing menu was difficult to use once too many apps registered as sharing targets. The new version handles long lists of sharing targets easily, and even offers a very good MRU interface.
System settings are available from the notification tray.
This makes turning WiFi off and on, among other things, much easier to do.
The Market is available in the corner of the app tray.
This makes getting to the market easier.
Widgets are available as a tab in the app tray.
This allows you to preview what widgets look like before adding them.
Scrollable widgets are handy.

In general, I'd say the experience was negative, and I'd recommend avoiding this upgrade until some of the problems are addressed.

Some quick thoughts prompted by SOPA/PIPA

2012-01-06T13:20:00.000-08:00

Neither side has agreed on the first principals, and both are trying to control the framing of the discussion.

For example, starting by dismissing those that do not agree there is a legitimate “intellectual property” (artificial scarcity) issue in the first place.

Personally, I don’t think there’s any possibility of compromise on the atomic issue of whether to regulate speech. Enforcing personal, not-for-profit copyright infringement will always conflict with freedom of speech. We’ve seen it in Russia silencing dissidents under the cover of piracy raids, and we’ve seen enemies and competitors trying to put each other on child porn filter lists for spite or market advantage in Scandinavia, and we’ve seen it with Universal vs. Megaupload here. Once the infrastructure is built, for whatever noble reasons, it will invariably be abused.

(I come from a time when copyright was a civil matter between companies, and comedians made fun of the FBI warnings at the beginning of video tapes.)

And, really, does a busker chasing down each and every sidewalk pedestrian make sense anyway? Artists will eventually adjust to the fact that they are buskers again. It seems to have worked for Cory Doctorow, Jonathan Coulton, and Louis CK.

PowerShell Wait-DirectoryChange FileSystemWatcher function

2012-01-05T16:01:00.000-08:00

Here's a handy way to watch for changes in a directory using the FileSystemWatcher in PowerShell.

function Wait-DirectoryChange
{
<#
.Synopsis
Listens to the file system change notifications and raises events when a directory, or file in a directory, changes.
.Parameter Directory
The directory to monitor, in standard or Universal Naming Convention (UNC) notation. Default is current directory.
.Parameter ChangeType
The type of change to watch for: Created, Deleted, Changed, Renamed, or All. Default is All.
.Parameter Include
The type of files to watch. For example, "*.txt" watches for changes to all text files. Default is all files.
.Parameter Timeout
The time to wait before timing out. Default is forever.
#>
[CmdletBinding()] Param
(
[string][ValidateScript({Test-Path $_ -PathType Container})] $Directory = $PWD.ProviderPath,
[IO.WatcherChangeTypes] $ChangeType = 'All',
[string] $Include,
[TimeSpan] $Timeout = 0
)
$watcher =
 if($Include) { New-Object IO.FileSystemWatcher $Directory,$Include }
 else { New-Object IO.FileSystemWatcher $Directory }
if($Timeout.TotalMilliseconds) { $watcher.WaitForChanged( $ChangeType, $Timeout.TotalMilliseconds ) }
else { $watcher.WaitForChanged( $ChangeType ) }
}
nal wdir Wait-DirectoryChange

The "Moderated" Response to IEBlog's Anti-Auto-Update Screed

2011-06-28T08:08:00.000-07:00

I submitted a reply to A Browser for All Windows Customers: it’s about and, not or - IEBlog - Site Home - MSDN Blogs, but it was "moderated" away. Apparently, Microsoft only approves comments by sycophants and strawmen.

Here is the comment, for anyone curious:

So foot-dragging is a feature?
That just seems like a rationalization for not supporting standards in the coming years (HTML5 inputs, e.g.), and being the holdout on auto-upgrades, the only hope many web developers had of seeing reasonable standards support (as opposed to kludgy workarounds) before the end of their career.
Hopefully Chrome Frame will make some inroads for the sake of developers, upon whose back the "great" IE experience comes. (Can someone from Microsoft supply, from their vast compatibility stats, how many top Internet sites are likely to use hardware rendering vs. how many take date inputs?)
Don't expect a lot of sympathy just because the job is impossibly big. Microsoft used anti-competitive practices to make IE the default browser, and has now found itself in the position of having to support up to FOUR rendering engines per browser per OS per service pack. And it's only going to grow, this thankless position of "infrastructure". More rendering engines will accumulate geometrically year after year (in a product that doesn't even show a profit) until Microsoft just cannot sustain any more and has to make some tough decisions about backward compatibility.
I think the track record shows that this realization often comes quite late in Redmond.

And another "moderated" comment, from The Perils of User-Agent Sniffing, 2011 Edition - EricLaw's IEInternals - Site Home - MSDN Blogs:

I'm surprised and disappointed that Google is using UA sniffing, but do you really expect all websites to understand the dozens of different combinations of rendering engine modes, browser modes, and document modes, and the endless HTTP extensions (X-XSS-Protection, DNT, X-UA-Compatible, X-Content-Type-Options, X-Frame-Options) required to get IE working?
Can anyone really be blamed for wanting to just walk away from IE, given that we've all just been slapped in the face again with the assertion that the web may only progress in 10-year increments?

F# Source that runs interactively or compiles

2011-05-23T12:06:00.000-07:00

Here's a simple way to create code that you can run with FSI or compile with FSC:

#if INTERACTIVE
// any #load, #r, or other fsi directives, e.g.:
#r "System.Configuration.dll"
#else
module ModuleName
#endif

// … code …

let main (args:string[]) =
    // … main body …

#if INTERACTIVE
main fsi.CommandLineArgs.[1..]
#else
[<EntryPoint>]
let Main args =
    try
        main args ; 0
    with
        | ex -> printfn "%s" (string ex) ; 1
#endif

Name the file with a .fsx extension, then you can compile it normally, or run it as a script with fsi filename.fsx -- args, or #load it from within FSI!

An F# mutually-tail-recursive CSV record parser.

2011-04-07T13:58:00.000-07:00

I've been really enjoying F# lately. Functional, concise, fast, fun. Here's a simple CSV parser. It's meant to be used ad-hoc, rather than some of the more full-featured CSV parsers that require what I'd consider to be an unrealistic level of premeditation to set up the columns and datatypes at design time. This one is a simple sequence of lists of, or mapped headers to, string values.

open System
open System.IO

// An F# mutually-tail-recursive CSV record parser.
// See the spec at http://creativyst.com/Doc/Articles/CSV/CSV01.htm
let rec csvrecord sep (tr:#TextReader) record (line:string) i =
  if i = line.Length then record @ [""]
  else
      match line.[i] with
      | '"' -> csvfield sep tr record "" line (i+1)
      | ' ' | '\t'-> csvrecord sep tr record line (i+1)
      |  c  when c = sep -> csvrecord sep tr (record @ [""]) line (i+1)
      | '=' when line.[i+1] = '"' -> csvfield sep tr record "" line (i+2) // Excel compatibility
      | _ -> // unquoted field data
          let fs = line.IndexOf(sep,i)
          if fs = -1 then record @ [line.Substring(i).TrimEnd()]
          else
              csvrecord sep tr (record @ [line.Substring(i,fs-i).TrimEnd()]) line (fs+1)
and csvfield sep (tr:#TextReader) record field (line:string) i =
  if i = line.Length then csvfield sep tr record (field+"\n") (tr.ReadLine()) 0
  elif line.[i] <> '"' then
      let q = line.IndexOf('"',i)
      if q = -1 then csvfield sep tr record (field+line.Substring(i)+"\n") (tr.ReadLine()) 0
      else csvfield sep tr record (field+line.Substring(i,q-i)) line q
  elif i = line.Length-1 then record @ [field]
  elif line.[i+1] = '"' then csvfield sep tr record (field+"\"") line (i+2)
  elif line.[i+1] = sep then csvrecord sep tr (record @ [field]) line (i+2)
  else // not an escaped quote and not end of field; try to recover by appending trimmed unquoted field data
      let fs = line.IndexOf(sep,i+1)
      if fs = -1 then record @ [field+line.Substring(i).TrimEnd()]
      else csvrecord sep tr (record @ [field+line.Substring(i,fs-i).TrimEnd()]) line (fs+1)

let csvrows sep (filepath:string) = seq {
  use sr = new StreamReader(filepath)
  while not sr.EndOfStream do
      yield csvrecord sep sr [] (sr.ReadLine()) 0
}

let csvrecords sep (filepath:string) = seq {
  let lists = csvrows sep filepath
  let headers = Seq.head lists
  for vals in (Seq.skip 1 lists) do
      yield List.zip headers vals |> Map.ofList
}

Then What? elif, elsif, elseif, or else if?

2011-02-16T10:54:00.000-08:00

"Else if" in multiple languages:

elif : Python, bash, and F#
elsif : Perl and Ruby
elseif : PHP and PowerShell
else if : C, C++, C#, JavaScript, and just about everything else (if)
ELSE IF : T-SQL (case-insensitive)
ElseIf : VB, VBScript, and VB.NET (case-insensitive)

Conspicuous Features Missing From Google Chrome

2010-12-15T14:27:00.001-08:00

Things that I miss when I use Google's Chrome browser:

feed subscription icons (I'll have to turn it on in Firefox 4, apparently)
feed display (sometimes I just want to download an enclosure, or see what's in a feed)
view selection source
full page info: vs.
about:config
feed-populated live bookmarks
dynamic smart bookmarks
bookmarks supporting dynamic microsummary live titles
functional Make Link extension (that copies as HTML)
Down Them All extension ("Right now there is no way to port dTa over without losing the majority of functionality.")
full Firebug

Things that I miss when I use Firefox:

speed!

Things that I miss when I use Internet Explorer:

not using Internet Explorer

US Federal Holidays in PowerShell

2010-12-11T12:33:00.000-08:00

Here's how to determine whether a date is a US federal holiday using PowerShell.

function Test-USFederalHoliday
{
<#
.Synopsis
Returns the name of the holiday of a date, if it is a U.S. federal holiday.
.Description
The following holidays are checked:
* New Year's Day, January 1 (± 1 day, if observed)
* Martin Luther King, Jr. Day, Third Monday in January
* President's Day, Third Monday in February
* Memorial Day, Last Monday in May
* Independence Day, July 4 (± 1 day, if observed)
* Labor Day, First Monday in September
* Columbus Day, Second Monday in October
* Veteran's Day, November 11 (±1 day, if observed)
* Thanksgiving Day, Fourth Thursday in November
* Christmas Day, December 25 (±1 day, if observed)
#>
[CmdletBinding()]
Param
(
[Parameter(Mandatory=$true)][datetime]$Date,
[Parameter(Mandatory=$false,HelpMessage='Are Saturday holidays observed on Friday?')][switch]$SatToFri,
[Parameter(Mandatory=$false,HelpMessage='Are Sunday holidays observed on Monday?')][switch]$SunToMon
)
$MMdd= '{0:MMdd}' -f $Date
switch($Date.DayOfWeek)
{
'Monday'
{
 switch -regex ($MMdd)
 {
  '^01(?:1[5-9]|2[01])$' {return 'Martin Luther King, Jr. Day'}
  '^02(?:1[5-9]|2[01])$' {return 'Presidents Day'}
  '^05(?:2[5-9]|3[01])$' {return 'Memorial Day'}
  '^090[1-7]$' {return 'Labor Day'}
  '^10(?:0[89]|1[01-4])$' {return 'Columbus Day'}
 }
 if($SunToMon)
 {
  switch($MMdd)
  {
   '0102' {return 'New Year''s Day (Observed)'}
   '0705' {return 'Independence Day (Observed)'}
   '1112' {return 'Veteran''s Day (Observed)'}
   '1226' {return 'Christmas Day (Observed)'}
  }
 }
}
'Thursday'
{
 if($MMdd -match '^112[2-8]$') {return 'Thanksgiving Day'}
}
'Friday'
{
 if($SatToFri)
 {
  switch($MMdd)
  {
   '1231' {return 'New Year''s Day (Observed)'}
   '0703' {return 'Independence Day (Observed)'}
   '1110' {return 'Veteran''s Day (Observed)'}
   '1224' {return 'Christmas Day (Observed)'}
  }
 }
}
}
switch($MMdd)
{
'0101' {return 'New Year''s Day'}
'0704' {return 'Independence Day'}
'1111' {return 'Veteran''s Day'}
'1225' {return 'Christmas Day'}
}
}

US Federal Holidays in F#

2010-12-11T00:55:00.000-08:00

Here's how to determine whether a date is a US federal holiday using F#. You probably want to curry the first two boolean parameters with the appropriate settings (some departments observe Sunday holidays on Monday, and some observe Saturday holidays on Friday).

open System
let getHoliday sunToMon satToFri (date:DateTime) =
   match date.Month, date.Day, (date.Day-1)/7+1, date.DayOfWeek with
   | 1, 1, _, _ -> "New Year's Day"
   | 1, 2, _, DayOfWeek.Monday when sunToMon -> "New Year's Day (observed)"
   | 1, _, 3, DayOfWeek.Monday -> "Martin Luther King, Jr. Day"
   | 2, _, 3, DayOfWeek.Monday -> "Presidents Day"
   | 5, d, _, DayOfWeek.Monday when d >= 25 -> "Memorial Day"
   | 7, 3, _, DayOfWeek.Friday when satToFri -> "Independence Day (observed)"
   | 7, 4, _, _ -> "Independence Day"
   | 7, 5, _, DayOfWeek.Monday when sunToMon -> "Independence Day (observed)"
   | 9, _, 1, DayOfWeek.Monday -> "Labor Day"
   | 10, _, 2, DayOfWeek.Monday -> "Columbus Day"
   | 11, 10, _, DayOfWeek.Friday when satToFri -> "Veterans Day (observed)"
   | 11, 11, _, _ -> "Veterans Day"
   | 11, 12, _, DayOfWeek.Monday when sunToMon -> "Veterans Day (observed)"
   | 11, _, 4, DayOfWeek.Thursday -> "Thanksgiving Day"
   | 12, 24, _, DayOfWeek.Friday when satToFri -> "Christmas Day (observed)"
   | 12, 25, _, _ -> "Christmas Day"
   | 12, 26, _, DayOfWeek.Monday when sunToMon -> "Christmas Day (observed)"
   | 12, 31, _, DayOfWeek.Friday when satToFri -> "New Year's Day (observed)"
   | _ -> null

You can really see the power of active pattern matching for the Memorial Day case.

No NTFS Compression in .NET?

2010-11-22T10:57:00.000-08:00

There is no class in the .NET SDK to use (de)compress files using NTFS! Apparently, people just use interop, which seems like way too much overhead to me.

PowerShell Parameter Defaults in Get-Help -full

2010-11-18T09:17:00.000-08:00

"Default values do not appear in the parameter attribute table, even when they are defined in the function or script."

WTF? Then why is it even there‽

Where is the .NET support for XSLT 2.0?

2010-11-09T07:53:00.000-08:00

Copyfight: Maria Cantwell's response to my concerns

2010-10-31T14:52:00.000-07:00

Earlier this year, I sent my concerns about ACTA (and the general scorched-earth approach to copyright that puts the rights of Lady Gaga and the Transformers above consumer protections and even basic human rights) to some politicians. The feedback went something like this:

I'll be as brief as I can:
Please, NO damage awards based on decades-old, half-remembered, vastly inflated, and entirely unscientific extrapolation of a rough estimate of copyright losses due to fake tractor parts. That's where they currently come from. Seriously. Look it up.
Please, NO enforcement that accepts accusation as conviction. In this country we are presumed innocent.
Please, NO enforcement that restricts or constrains Internet access. The Internet is where we work, and play, and socialize, and seek medical advice and employment, and have political discussions, and provide feedback to elected officials. [France passed a "three-strikes" law, and is getting 25,000 accusations a day.] [Only 1.7% of sites blocked by Scandinavia's "child-porn" filters are actually child porn - Boing Boing]
Please, do not sacrifice free speech for corporate profits and Hollywood campaign donations.
Please, NO mandates for ISPs to snoop on their customers. That is an end-run around the Fourth Amendment. [In Russia, this has been a convenient way to silence dissidents.]
Please, protect fair use and punish blanket takedown notices and other baseless legal bullying.
Please, ensure that groups purporting to collect royalties for artists are actually paying the artists.
Please, ensure that companies externalizing the cost of enforcement of their copyrights to the government are actually paying their taxes.
Please, stop referring to unauthorized copying as theft (theft deprives the original owner of something), or piracy (piracy is a very real life-threatening issue that should not be trivialized).
Please, NO more pressuring other countries to accept our laws on their soil.
Please, NO more demonizing Canada for specious copyright transgressions.
Please, accept that DRM is never going to work technically, and only serves to frustrate legal purchasers of content.
Please, stop bundling music and movies with pharmaceuticals. Using fear about counterfeit drugs is a cheap tactic.
Please, consult with Cory Doctorow http://craphound.com/ , the EFF http://www.eff.org/ , and the CDT http://www.cdt.org/ .
Personally, I'd like to see us return to a pre-DMCA world where copyright was a civil matter. Sadly, I've scaled back my hopes significantly, and I now just don't want you to destroy freedom to protect a specific business model.
Thank you for your time.

What I got back from Maria Cantwell in July was very disappointing.

Dear Mr. Lalonde,
Thank you for contacting me regarding the Anti-Counterfeiting Trade Agreement (ACTA) negotiations. I appreciate hearing from you on this matter and sincerely regret the delayed response.
As you may know, the ACTA negotiations began in June 2008 with the participation of Australia, Canada, the European Union and its 27 member states, Japan, Mexico, Morcocco, New Zealand, Republic of Korea, Singapore, Switzerland, and the United States. The negotiations, which are scheduled to conclude in 2010, have the goal of providing an international framework to improve enforcement of intellectual property right (IPR) law.
The ACTA initiative aims to create improved, international standards to fight the growing problem of counterfeiting and piracy. The effort is not intended to interfere with the fundamental rights and civil liberties of ordinary citizens, and will be consistent with the WTO Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) and the Declaration on TRIPS and Public Health.
I am aware of your concerns about the transparency of the ACTA negotiations. However, it is an accepted practice during trade negotiations not to disclose negotiating texts to the public thus enabling officials to engage in the frank exchanges necessary for agreement on complex issues.
You will be pleased to know that the USTR has taken steps to ensure an unprecedented level of transparency in the ACTA negotiations, and won endorsement of the importance of meaningful public input at the second level of negotiations in November 2009. Additionally, on April 21, 2010, the USTR released a public, pre-decisional draft text of the ACTA. It is available for your reference through the USTR website at http://www.ustr.gov/acta.
Thank you again for contacting me to share your thoughts on this matter. You may also be interested in signing up for periodic updates for Washington State residents. If you are interested in subscribing to this update, please visit my website at http://cantwell.senate.gov. Please do not hesitate to contact me in the future if I can be of further assistance.
Sincerely,
Maria Cantwell
United States Senator

I've replied, but don't expect to hear back.

I have to say that I'm am deeply disappointed.
By bringing up public health, it seems clear that you are among those trying to conflate physical counterfeiting with unauthorized digital copying. "Consistency" is also an obvious ploy to slowly ratchet up onerous copyright law by getting excessive laws passed in "sympathetic" regimes then demanding them here.
And far from ensuring unprecedented levels of transparency, it is clear that the USTR is doing everything in his power to keep these negotiations, and the current treaty itself, secret. This is well documented.
http://boingboing.net/2010/02/25/acta-leak-now-we-kno.html
As I rejected it when Vice President Cheney attempted to claim that secrecy was necessary to have "frank discussions" in his energy policy discussions, I reject it here. Anything that can't be said in public is shady when developing policy in a democracy.
The text of ACTA has once again leaked, and once again, it is clear that its supporters (yourself included, apparently) will not rest until all of our Internet traffic is monitored (allegedly for "infringing content"), our laptops and iPods are searched at the border, and entire families are collectively punished (counter to Genevea Convention "consistency") based on specious accusations of media companies.
http://www.boingboing.net/2010/07/14/acta-leaks----again.html
Please tell me you have the integrity to dismiss the made-up losses media companies are constantly parroting, to allow us to create without the constant fear of corporate legal threat, and to protect our right to speak and assemble digitally for the future.
http://www.fsf.org/campaigns/acta/acta-declaration/

Perhaps not surprisingly, Senator Cantwell's second-highest contributor is Microsoft.

Mini Review: Swype and SlideIt Android sliding keyboards

2010-10-24T09:53:00.000-07:00

Sliding keyboards work amazingly well. Given a moderately practiced user typing English text and reasonably fast hardware, I suspect that a "typing" speed could be achieved that would be nearly competitive with a full-sized mechanical keyboard.

Android provides at least two sliding keyboards so far, and here are their relative strengths and weaknesses.

SlideIt

Fastest for raw text, though this difference may only be noticeable on lower-end hardware.
Supports custom macros, called "shortcuts" that allow for arbitrary letter combinations to expand to any arbitrary text.
Does not require swiping over the apostrophe for contractions.
Switching to the symbols keyboard then sometimes requires scrolling through rows of symbols, and automatically switches back to the regular keyboard after typing each symbol (which can be frustrating when typing pairs of parentheses or brackets), unless "pinned", which requires an additional keystroke.
Symbols and accented letter keys are visible and can be found easily (if not quickly) when scrolling through keyboard.
Has built-in voice input support. This is mostly for Android 1.x devices that do not already have this support.
Has built-in Graffiti (Palm-style block-letter recognition) support.
Limited paged-horizontal display of alternative word matches.
Irritatingly disables the shift key after the first letter of a word is typed.
More than twenty languages supported.
Always includes a return key.
Unique keys: ¬ …

Swype

Larger keys.
Much faster support for capitalization by swiping a letter up above the keyboard, then to the next letter. This makes typing "I" in particular a much faster upward flick than the more cumbersome [Shift], [i].
Much faster support for symbols (by long-pressing a letter for common symbols, or switching to the symbols keyboard for a more comprehensive selection).
Extended symbols and accented characters are not visible until long-long-pressing a key (in either symbol or regular keyboard mode). Location of symbols can be difficult to find and remember until the mnemonically understood.
Numeric keyboard.
Limited paged-vertical display of alternative word matches.
Contractions require swiping the apostrophe.
Swiping period/exclamation/question to space quickly ends a sentence.
Variable speed-vs.-accuracy slider setting.
Extra hidden keyboard for powerful cursor, selection, and clipboard control.
Three languages supported (English, French, Spanish).
Return key is often irritatingly replaced with "Done" key.
Unique keys: ¿ ¡ „ ð · þ ¶ ª æ œ ß

The New Caste System

2010-09-15T09:18:00.000-07:00

Welcome to the new caste system. You are either a Creator (noble) or a Consumer (pleb/serf).

Plebs don't get to own anything, like the music, movies, books, or software they buy; they may only license what they purchase. No one without an existing defensive copyright/patent portfolio will be able to become a Creator, since "creation" is only achieved by transforming or combining existing ideas.

Slowly, the scope will creep as you are required to pay a periodic license fee for the architecture you live in, the design of the furniture you use, the clothing and shoes you wear, &c.

Punishment for unauthorized copying/use ("piracy") will just get more draconian, initially as entire households are excommunicated from the Internet for unsubstantiated accusation by the corporate scanbots, but ultimately as "intellectual" "property" (imaginary property) enforcement provides an excuse for capricious, arbitrary, and unlimited affronts to legal rights of protestors and dissidents.

So, anyway, happy Wednesday.

Quixotic IE6 Workarounds

2010-04-06T08:19:00.000-07:00

Microsoft has renewed support for IE6 until 2014.

If this bothers you, be sure to remember this the next time you are preparing to give Microsoft money!

To celebrate this spineless irresponsibility, here are some coping strategies:

DENG, perhaps the first such major attempt to work around IE6, is a Flash-based standards-capable browser. Sadly, it appears to have been abandoned years ago.
Dean Edwards' IE7 is a JavaScript library to fix CSS rendering in IE6.
IE6 CSS Fixer is an online tool that uses a list of known tricks to update CSS with IE6 hacks and workarounds.
Push Up the Web has a subtle upgrade link script for users of outdated browsers, including IE6.
IE6 Update is a warning to encourage IE6 users to upgrade that looks like the IE infobar.
IE6 No More is a bigger warning to encourage IE6 users to upgrade.
sevenup and ie6-upgrade-warning are large, attractive warning messages to encourage IE6 users to upgrade or install another browser.
There are several WordPress plugins and techniques to discourage IE6: Shockingly Big IE6 Warning, Serve IE6 Visitors the Default WordPress Theme, IE6 No More, Block IE6, Anti IE6, Kill IE6.
Google Chrome Frame hosts Google's Chrome browser within IE6 as a plugin that installs with a minimum of effort.
Edskes Software Silent Setup for Mozilla Firefox is a minimal-friction installer for Firefox.

It must be said, however, that if IE6 is a small (single digit percentage) of your web traffic, you probably needn't to spend as much extra time on it as you have in the past. Most things will still work well enough if you build for progressive enhancement/graceful degredation anyway, even though sites may not look quite as beautiful in IE6 without massively increasing your development time to implement various workarounds.

Is the end of IE6 finally in sight?

2010-02-24T20:47:00.000-08:00

Internet Explorer Contract Addendum - Monday By Noon • Should I bother supporting IE6? - Pro Webmasters - Stack Exchange • Browse Sad • RIPIE6.com • IE6 No More • Bring Down IE 6 • Explorer Destroyer • I Don't Support IE6 • I dropped Internet Explorer 6 • StopIE • Stop Internet Explorer • SaveIE6 ;) • Dear IE6 • IE6ify Bookmarklet • IE sucks • Why IE Sucks • Enough IE6 • Is Internet Explorer 6 Dead? • Do I have IE6? • IE6 MUST DIE • Death To IE6 • IE Death March • HEY-IT - We want to get rid of IE6! • Explorer Exposed! • IE6 Must Die - Support now! - Twibbon.com • Phase Out IE6 • IE6 Update • DearIE6 - So Long • GOODBYE IE6... I hate you! • IE6 Redirection Script • IE 6 Blocker Script | CSS-Tricks • IE6 Countdown to Death • Too Cool for Internet Explorer • Update Your Browser • Browser-Update.org • Browser Upgrade Information • Pushuptheweb.com • Kill Bill's Browser - Switch to Firefox • Browse Happy • It's Time to Choose A Better Browser • Alternative Browser Alliance • Crash IE • IEUpdate.nl • drop-ie6.dk — Sig farvel til Internet Explorer 6 • ie6 - aktualizuj już teraz! • IE6 Funeral • ~~KILL IE 6~~ • ~~NoIE6~~ • ~~IE6: Do NOT want!~~ • ~~The best way to support IE6?~~ • ~~Am I using IE6?~~ • ~~End 6!~~ • ~~StopIE6.com~~ • ~~IE6Offenders~~ • ~~The IE6 List~~ • ~~Stop Living in the Past! Upgrade IE6~~

Word on the street is that July 13, 2010 is the official end of life for IE6. Maybe it'll get extended, but Digg has already dropped support, Google Sites and Docs will on March 1, 2010, Google has announced YouTube will drop support March 13, 2010, and Microsoft has been encouraging users to upgrade.

There have been efforts toward, and hopes and predictions of IE6's death in the past, and maybe some (primarily corporate) users may continue to use it anyway. However, Google provides huge momentum, and news coverage, discussion, angry T-shirts, and voodoo dolls are going to make justification of coasting along on IE6 much harder from now on. Norway has been in the fight for a while, there's a petition to pull IE6 from the UK government, and even the hubris of a funeral for IE6.

It just seems like the time is right to spend less time on crazy, quixotic fixes, and invest some time (at least for a while) helping educate or encourage users to make a better browser decision.

Flickr: John Martz's Photostream

Converting between XDocument and XmlDocument, XElement and XmlElement

2010-02-01T13:02:00.000-08:00

Here's a class to provide extension methods for conversion between System.Xml and System.Xml.Linq classes:

using System.Xml;
using System.Xml.Linq;

/// <summary>
/// Provides extension methods for simple conversion between System.Xml and System.Xml.Linq classes.
/// </summary>
public static class XmlLinqConversionExtensions
{
  /// <summary>
  /// Converts an XDocument to an XmlDocument.
  /// </summary>
  /// <param name="xdoc">The XDocument to convert.</param>
  /// <returns>The equivalent XmlDocument.</returns>
  public static XmlDocument ToXmlDocument(this XDocument xdoc)
  {
    var xmldoc = new XmlDocument();
    xmldoc.Load(xdoc.CreateReader());
    return xmldoc;
  }
  
  /// <summary>
  /// Converts an XmlDocument to an XDocument.
  /// </summary>
  /// <param name="xmldoc">The XmlDocument to convert.</param>
  /// <returns>The equivalent XDocument.</returns>
  public static XDocument ToXDocument(this XmlDocument xmldoc)
  {
    return XDocument.Load(xmldoc.CreateNavigator().ReadSubtree());
  }
  
  /// <summary>
  /// Converts an XElement to an XmlElement.
  /// </summary>
  /// <param name="xelement">The XElement to convert.</param>
  /// <returns>The equivalent XmlElement.</returns>
  public static XmlElement ToXmlElement(this XElement xelement)
  {
    return new XmlDocument().ReadNode(xelement.CreateReader()) as XmlElement;
  }
  
  /// <summary>
  /// Converts an XmlElement to an XElement.
  /// </summary>
  /// <param name="xmlelement">The XmlElement to convert.</param>
  /// <returns>The equivalent XElement.</returns>
  public static XElement ToXElement(this XmlElement xmlelement)
  {
    return XElement.Load(xmlelement.CreateNavigator().ReadSubtree());
  }
}

What's in your (Windows) path?

2009-11-17T16:03:00.000-08:00

perl -le "@x=split/;/,$ENV{PATHEXT};for(split/;/,$ENV{PATH}){print;next unless chdir$_;print qq'\t'.join(qq'\n\t',map{<*$_>}@x)}"

This one-liner will list all executables currently in your path, so you can decide if maybe you want to trim that path a bit, maybe by transferring a few things to doskey macros.

If you add doskey macros, be sure to add a new string value to HKEY_CURRENT_USER\Software\Microsoft\Command Processor called Autorun with the data C:\WINDOWS\system32\doskey.exe /macrofile=macrofile.

Netflix Vista Media Center Plugin: working again

2009-11-11T21:05:00.000-08:00

(yay)

Netflix Vista Media Center Plugin Broken. Again.

2009-11-06T23:19:00.000-08:00

I wonder if the plugin now only supports the brand-new Windows 7, or if this is just buggy, buggy code. Was this broken by one of the near-daily patches Vista has been insisting on lately? Should I start second-guessing Microsoft patches, in case they downgrade functionality, similar to the way iTunes does?

Without Netflix in VMC, I have to decide whether to go back to Mythbuntu (auto commercial skip and better music playlists) or not. It's a shame after all those reinstalls it took to get the 'mature' Vista OS working. But then again, it'd be nice to be able to print again—my print server, which works fine under Linux, is no longer supported under Vista, so I'm supposed to replace it, even though it works fine. Not to mention that not one of my dozen or so gaming controllers works under Vista (or 7, for that matter).

My current media center

2009-09-10T22:02:00.000-07:00

So I have been forced, against my will, to install Windows on my television PC in order to see the Netflix online content I pay for.

I've fought countless Windows bluescreens in the process of setting this up (more on those in a later post), been frustrated by Windows Update failing and leaving my system in an un-update-able state, and been puzzled by Windows' consistent inability to recognize my network devices on any home computer I've set up in the last decade (I always end up booting into Linux to get the network drivers to spoon-feed to Windows). I even found out that switching the Media Center theme to "high visibility" irrevocably breaks the "Internet TV beta".

I have, however, come up with a decent list of software if you must put Windows on your TV:

Windows Media Center (it probably came with your Windows edition)
the Netflix plugin for WMC (the logo should be in WMC already, and will link to the download site)
Boxee alpha, a neat open-source media center with scores of plug-ins full of content (it's cross-platform if you don't have to use Windows)
hulu Desktop, another media center, just for hulu content (hulu hates Boxee for some reason, and must not have signed a multi-million dollar deal to become a Microsoft parner to show up in Media Center by default)
Steam, the iTunes of computer games
Google Earth just looks very cool on a big TV

Highlighting Text or Fields in Windows

2009-08-18T10:26:00.000-07:00

How does Windows always incorrectly guess whether I want a field (such as a browser address bar) fully selected when I click it? It must use a similar algorithm to how it incorrectly judges when to suddenly highlight entire words when that is not my intention.

This is what I get for not using the keyboard exclusively, I guess.

Why are people still using IE6?

2009-08-11T08:20:00.000-07:00

Dean Hachamovitch has an article, IEBlog : Engineering POV: IE6, on the IEBlog, about IE6's continued use. Basically, many IE6 users think it works fine for them, and don't see a reason to upgrade (note to malware authors: these people and IT departments should be easy exploits for some time yet).

Yes, keep "users in control of their PCs" (I assume this means Microsoft will allow me to uninstall their DRM layer in Windows 7?).

This does not mean people should have the expectation that the entire Internet should support Netscape 1.0, Lynx, WebTV 1.0, or MSIE 6. The Internet evolves (as do the security threats). If users want a static environment, they should go pick up a CD-ROM encyclopedia from 1992.

Sure, if a company wants to cater to the IE6 audience, they should spend the time, but I'd bet that someone unwilling or unable to upgrade to a secure web browser isn't as likely to be profitable enough to justify all of the extra work that requires.

I do have to observe that since Microsoft has lived by inertia, the reason IE usage grew for so long, and the reason Windows has utterly dominated for so many years, that it's refreshing to see them cursed by inertia, having to support Windows XP and IE6 for years to come.

Microsoft won't be able to kill XP/IE6 any more than the financial industry has been able to completely kill COBOL.

Internet Explorer User Agent Spam

2009-07-28T10:00:00.000-07:00

Update: No more! IEBlog : Introducing IE9’s User Agent String

Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; GTB6; Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1) ; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; OfficeLiveConnector.1.3; OfficeLivePatch.0.0)
Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0; Trident/4.0; Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1) ; Embedded Web Browser from: http://bsalsa.com/; SLCC1; .NET CLR 2.0.50727; Media Center PC 5.0; InfoPath.2; .NET CLR 1.1.4322; .NET CLR 3.5.30729; .NET CLR 3.0.30618; OfficeLiveConnector.1.3; OfficeLivePatch.0.0)
Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; .NET CLR 1.0.3705; .NET CLR 1.1.4322; Media Center PC 4.0; .NET CLR 2.0.50727; IEMB3; Windows-Media-Player/10.00.00.3990; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; eMusic DLM/3; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; WWTClient2; OfficeLiveConnector.1.3; OfficeLivePatch.0.0; IEMB3; AskTB5.3)
Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; Trident/4.0; Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1) ; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; MSN Optimized;US; .NET CLR 3.0.04506.648; Zune 3.0; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; MSN Optimized;US; AskTB5.4)
Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; Comcast Install 1.0; .NET CLR 1.0.3705; .NET CLR 1.1.4322; Media Center PC 4.0; Media Center PC 3.0; .NET CLR 2.0.50727; .NET CLR 3.0.04506.30; .NET CLR 3.0.04506.648; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; OfficeLiveConnector.1.3; OfficeLivePatch.0.0)

Why are the .NET CLR details cluttering up the user agent string? I can't figure out how the information is relevant in the web environment.

We've got user agent strings pushing past FOUR HUNDRED CHARACTERS in length.

If Microsoft is trying to communicate browser capabilities, there's a header for that: HTTP Accept.

The same goes for Zune, MSN Optimized, Media Center PC, InfoPath, OfficeLiveConnector, OfficeLivePatch, and MSN Optimized, and probably dozens of others.

Microsoft is filling up our logs, and congesting the tubes with what can only be considered SPAM.

My web logs are not the place for Microsoft advertising!

Also: It's past time to get rid of the "Mozilla/4.0 (compatible;" nonsense.

I propose a protest: include the three longest MSIE user agent strings from your server logs at the top of all comments and communications to Microsoft.

Kindle and DRM

2009-07-17T19:16:00.000-07:00

Web-based UML diagramming apps

2009-07-07T10:07:00.000-07:00

If you need to quickly create a UML diagram, I've found a couple of sites provide a simple way to convert simple text representations into nice, attractive UML diagrams.

WebSequenceDiagrams.com

This site is great for creating UML sequence diagrams.

Alice->Bob: Authentication Request
note right of Bob: Bob thinks about it.
Bob-->Alice: Authentication Response

yUML

yUML.me easily creates both class and use case UML diagrams.

[Customer]+1->*[Order]
[Order]++1-items >*[LineItem]
[Order]-0..1>[PaymentMethod]

[User]-(Login)
[User]-(Logout) 
(Login)<(Reminder) 
(Login)>(Captcha)

For more, see my diagram bookmarks.

Best Registry Hack Ever: MaximizeApps

2009-07-01T09:40:00.000-07:00

This registry hack maximizes windows by default when they open! For years, I've been using a superfluous freeware program for this. I wish I'd known about this much sooner, or that Microsoft had provided some UI for this option (at least in the old TweakUI).

Windows Registry Editor Version 5.00

[HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer]
"MaximizeApps"=dword:00000001

Update: Here's how to set it in PowerShell ( ` is the line-continuation character):

New-ItemProperty HKCU:\Software\Microsoft\Windows\CurrentVersion\Explorer `
-Name MaximizeApps -Value 1 -PropertyType DWORD

Easy XML serialization of complex objects with C# 3.0

2009-06-10T11:19:00.000-07:00

C# 3.0 Update! If you've got complex, XML-serializable objects you'd like to trace easily, such as SOAP objects, here's a handy way to dump their contents to an XML string:

using System.IO;
using System.Text;
using System.Xml;
using System.Xml.Serialization;

public class XmlExporter
{
  /// <summary>
  /// Serialize an object as an XML string.
  /// </summary>
  /// <param name="o">The object to serialize.</param>
  /// <param name="defaultNamespace">The default XML namespace of the object (to supress noisy namspaces).</param>
  /// <returns>The object, serialized as an XML string.</returns>
  public static string ToString(object o, string defaultNamespace)
  {
    if (o == null) return "<NULL />";
    var utf8 = new UTF8Encoding(false);
    using (var ms = new MemoryStream())
    using (var x = new XmlTextWriter(ms, utf8) {Formatting = Formatting.Indented})
    {
      var xs = String.IsNullOrEmpty(defaultNamespace)
        ? new XmlSerializer(o.GetType())
        : new XmlSerializer(o.GetType(), defaultNamespace);
      xs.Serialize(x, o);
      return utf8.GetString(ms.GetBuffer(), 0, (int) ms.Length);
    }
  }
}

Lye Service: A SOAP service that uses XML strings rather than objects

2009-06-09T08:38:00.000-07:00

I've decided on a name for SOAP services that use XML strings for input and output rather than objects: Lye Services.

These services completely miss the point of the SOAP overhead, which is to hide the obnoxious XML string manipulation in the SOAP layer, to avoid the tedious and error-prone process of building XML strings manually via concatenation or templates. This means you get all of the drawbacks of SOAP (high network overhead, heavy library dependence, XML serialization load) without any of the benefit (a complete service description; straightforward updating; working with objects and not worrying about XML strings, serialization, de-serialization, and validation).

The result is a service definition that is not complete: the schema and documentation for the XML to input is elsewhere, usually out-of-band, that does not automatically update when the service is updated.

Worse still, these POX-y services often omit a schema for the input XML that could be used to generate or validate the XML before sending it to the service, making the process much more error-prone. Sometimes the service is so poorly specified or documented that the developer is forced into tedious trial and error to figure out the appropriate input.

Thus, "Lye", an archaic and harsh soap that homophonetically implies an untrue implementation.

When encountering a Lye Service, it's worth asking whether a more structured approach (objects, rather than XML strings) could be used. If not, REST may be a better fit than SOAP.

Remembering IE databinding

2009-06-01T16:16:00.000-07:00

I was recently thinking about Internet Explorer 4's HTML data binding support, and it occurs to me that it is similar, at least superficially, to ASP.NET AJAX 4.0 declarative data controls. Where they differ tells a story about the difference between Microsoft 1997 and Microsoft 2009.

Cross-browser: MSAJAX doesn't demand IE loyalty, or pretend that other web browsers do not exist.
Generalized framework: Rather than play whack-a-mole with very specific features built in at the browser level, MSAJAX is extensible by ordinary web developers.
Security: Databinding had some issues it could never really overcome, in that it was essentially sending SQL straight from the client to the database server. MSAJAX naturally shows a great deal of maturity in anticipating security problems.
Standards: Microsoft, in recent years, clearly values standards more than it did back in the days of browser-war HTML-extensions, using normal XML/HTML/CSS2.1/REST to continue to advance the web platform.

IE8 has a find bar

2009-03-27T21:11:00.000-07:00

IE8 finally has a find bar like everyone else.

No CSS border-radius support, though. :(

Mean Date Measures

2009-03-23T12:34:00.000-07:00

365.2425 mean days per Gregorian year
30.436875 mean days per month
52.1775 mean weeks per year
4.348125 mean weeks per month
260.8875 mean weekdays per year
21.740625 mean weekdays per month
250.8875 mean US workdays per year (mean weekdays minus ten federal holidays)
20.90729166 mean US workdays per month

IE8 Accelerators

2009-03-19T11:20:00.000-07:00

Is it just me, or are IE8's Accelerators very similar to IE6's Smart Tags or context menu extensions? I guess it isn't just me.

IE8

2009-03-19T10:29:00.000-07:00

IE8 has...

an awesome bar, like Firefox
native, automatic color-coded tabs, like the Firefox extension
Chrome-style tab process isolation
CSS 2.1 support (full?)
"Accelerators"
Web Slices
images in search box suggestions ("Visual Search")

All of the sudden, Microsoft is talking a lot of smack about CSS 2.1 support in those other browsers.

VS2010 Web.config transforms

2009-03-19T10:24:00.000-07:00

Oh great, another XML transform language. Hopefully, everyone will make one suited to their niche, so knowledge won't be too portable. The reason? It's smaller. So I guess Microsoft will be switching all of their development efforts to APL or Perl, then?

T24F The Microsoft Web Sandbox

2009-03-18T13:10:00.000-07:00

first gen: Facebook JS (FBJS), AdSafe
second gen: Caja, FBJS2
ECMA TC-39 Security Working Group
provides W3C standard DOM support for even IE browsers
provides isolation for 3rd party scripts (3JS)
provides QoS checking for 3JS (only the embedded [div] portion fails or times out)
allows integration into rest of page (not absolute isolation)
JSVM script, 3JS transform service converts 3HTML or 3JS into a JSON closure that intercepts all DOM namespace lookups
virtual title and status bar for the hosting element
Apache License
well-formed HTML only
no document.write()
no eval()
no JS with statement
debug complexity
performance penalty

MIX09 Keynote: Scott Guthrie

2009-03-18T13:02:00.000-07:00

ASP.NET MVC 1.0
ASP.NET 4.0
VS 2010: parallel web.config files (Web.Test.config, Web.Release.config)
Microsoft Web Platform Installer
Windows Web App Installer/Gallery
Azure Services Platform
Silverlight 3: people are totally using it! (NetFlix, NBC Olympics), GPU hardware support, codec support (H.264, AAC, MPEG-4)
Expression Web 3
Expression Blend 3: SketchFlow, Photoshop/Illustrator import, source control
Eclipse support for Silverlight

Several guests, including Joel and Jeff from StackOverflow.com.

MIX09 Keynote: Bill Buxton

2009-03-18T11:37:00.000-07:00

"return on experience"
it's not the product, it's the experience"
the industrial designers from the 1920s are analogous to modern UX designers
Post-Its® for sequence design (like Dmitri Martin)
transitions are the most important part of state transition diagrams

StackOverflow podcast live at MIX09

2009-03-17T21:18:00.001-07:00

Joel and Jeff (left to right) discussing (in what should show up as SO podcast #46) data formats, the importance and peril of understanding technical historical context for programmers, and chopping off the ends of the meatloaf.

MIX09 at the Venetian

2009-03-17T21:00:00.000-07:00

Pardon the horrible quality of the photos. My "camera" is just barely a phone, much less a camera.

Yep, they dyed the "Grand Canal" at the Venetian green for St. Patrick's Day.

MIX09 swag (initial)

2009-03-17T14:06:00.000-07:00

Not pictured: a fairly innocuous black tote bag (a welcome change from the white-tarp-purses from last year), and a black T-shirt. No berets this year (yet).

Will Microsoft release IE8 at MIX09? | Hubdub

TSA: They don't know, either

2009-03-15T10:38:00.001-07:00

A warm, moist USB coccoon

2009-03-03T07:53:00.001-08:00

Mummify yourself with USB warmers:

Plus, there seem to be a huge amount of USB mug warmers:

It's almost as if sticking a heating element on a 5V power supply isn't all that difficult.

Also, there must be quite a few cold offices out there.

New MCE Remote: nmediapc HTPCKB

2009-02-27T11:44:00.000-08:00

My new 2.4GHz RF Wireless Keyboard with Track Ball & Remote Combo Set is intended for a Windows MCE system, but my intent is to get it working with my MythTV setup. So far, only a few buttons on the remote work with Mythbuntu (the keyboard is just a wireless USB keyboard), but here are the details I've gathered so far:

Button	showkey	showkey -s	xev
☒ Close	56+62	`38 36 B8 BE`	(closes xev)
Power	142	`E0 5F E0 DF`
My TV
My Music
My Picture [sic]
My Video
Live TV
Recorded TV
Guide	362
DVD Memu [sic]
MCE Button	56,125,28	`38 B8 E0 5B E0 DB 1C 9C`	Alt_L,Super_L,Return
Volume +	115	`E0 30 E0 B0`
Volume -	114	`E0 2E E0 AE`
Channel +	104	`E0 49 E0 C9`	Prior
Channel -	109	`E0 51 E0 D1`	Next
OK / Enter	28	`1C 9C`	Return
← Back / Clear	14	`0E 8E`	BackSpace
i Info	130	`E0 06 E0 86`	SunProps
▲	103	`E0 48 E0 C8`	Up
►	106	`E0 4D E0 CD`	Right
▼	108	`E0 50 E0 D0`	Down
◄	105	`E0 4B E0 CB`	Left
○ Record	29+19	`1D 13 9D 93`	Control_L+p
■ Stop	166	`E0 24 E0 A4`
▶/▮▮ Play/Pause	164	`E0 22 E0 A2`
◄◄ Rewind	168	`E0 18 E0 98`	XF86AudioRewind
►► Fast Forward	208	`E0 34 E0 B4`	XF86Forward
\|◂ Previous	29+48	`1D 30 9D B0`	Control_L+x
▸\| Next	163	`E0 19 E0 99`
Mute	113	`E0 20 E0 A0`
1	79	`4F CF`	KP_End
2	80	`50 D0`	KP_Down
3	81	`51 D1`	KP_Next
4	75	`4B CB`	KP_Left
5	76	`4C CC`	KP_Begin
6	77	`4D CD`	KP_Right
7	71	`47 C7`	KP_Home
8	72	`48 C8`	KP_Up
9	73	`49 C9`	KP_Prior
*	55	`37 B7`	*
0	82	`52 D2`	KP_Insert
#	42+4	`2A 04 AA 84`	#
Num Lock	69	`45 C5`	Num_Lock
Clear / ← Back	14	`0E 8E`	BackSpace
Enter / OK	28	`1C 9C`	Return

The keyboard and mouse are a step up from my previous setup, since they are both RF rather than IR, and the remote is considered a keyboard, so no lirc configuration is needed. Best of all, there is a trackball in the remote, so I don't need to drag out the keyboard to watch web TV.

Show your Google Calendar in your sidebar (without a separate file)

2009-02-04T10:20:00.000-08:00

Update: I guess it would just be easier to simply bookmark the URL in the src attribute after step 2, and mark that to load in the sidebar.

Inspired by the Lifehacker article Step By Step: Display Your GCal Agenda in Firefox's Sidebar, I've improved the process by using a data: URL instead of an external file.

In Google Calendar, go to Settings for your calendar, then click the "Customize the color, size, and other options" link to enter the calendar customization tool.
Choose "Agenda" View, and otherwise configure the calendar however you wish. Just leave the height and width at the default values.
Copy the HTML code.
Head over to The data: URI kitchen and paste in the HTML code.
Make the following changes:
- Check the "base 64" checkbox.
- Change src="// to src="http://
- Paste the following just after scrolling="no" and just before ></iframe>:
  onload="var c=this;c.style.height=window.innerHeight-16;c.style.width=window.innerWidth-16;c.onload=null;window.onresize=function(){c.style.height=window.innerHeight-16;c.style.width=window.innerWidth-16;}"
Click the Generate button.
Your calendar should appear, filling the page. Bookmark it.
In the Properties for this bookmark, check the "Load this bookmark in the sidebar" checkbox.

Now, you can just click the bookmark to open the Google Calendar in your sidebar!

Last minute gift idea: support a podcast with a donation

2008-12-23T17:40:00.000-08:00

Support a great podcast in someone else's name:

- TEKDIFF (teknikal diffikulties)- featuring The Account
Decoder Ring Theatre
Escape Pod and PodCastle
The Moth

…especially public radio, which seems to be the backbone of podcasting:

I don't read Russian

2008-10-28T07:23:00.000-07:00

A great deal of Cyrillic spam has made it into my various inboxes lately. I don't read Russian, so I'd sure like to be able to configure my email accounts to not accept stuff I cannot read, but this isn't an option‽

Configuring acceptable languages has been an underused but nonetheless available feature of HTTP for at least a dozen years. Why can't MTAs be configured to look for character encodings that the user cannot use? This would be a simple, straightforward solution to most of the spam that somehow currently makes it through my filters.

Update: Other languages I don't read include: Farsi, Chinese, Arabic, Korean, Georgian, and Japanese. The fault lies in myself, sure, but I need an MTA that can say Let me just stop you there. before an opaque message is relayed.

Rockbox podcast cover art

2008-10-21T22:15:00.000-07:00

The fastest way to add cover art to a podcast folder on an MP3 player running Rockbox:

Browse to the raw feed using Firefox
Press Ctrl+U
Type /:im (this finds the Apple iTunes image tag, the best image to use for the podcast)
Copy the URL of the iTunes cover art image
In a bash terminal window, in the podcast directory, type wget `xclip -o -d` ; convert *.jpg -resize 100x100 cover.bmp ; rm *.jpg

You'll need to be using Linux, and have xclip and convert installed, of course.

User Agent Web Of Lies: Where Does It End?

2008-09-03T07:25:00.000-07:00

Internet Explorer started it by lying in its user agent string, claiming to be Mozilla (meaning Netscape 4). It got worse with Opera claiming to be both Mozilla and MSIE. Safari included both Mozilla and Gecko (meaning Firefox, et al.) in its UA string. But now, Chrome mentions Mozilla, Gecko, and Safari/WebKit. Will someone please put a stop to this nonsense? Anyone still using a user agent string to select functionality deserves to be hung and drawn and quartered and whipped and boiled and chopped into little bits, then to have those bits jumped on. This snowballing user agent string business just serves to make HTTP slower, make web logs bigger, and parsing logs much more tedious.

Google Chrome User Agent String

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/525.13 (KHTML, like Gecko) Chrome/0.2.149.27 Safari/525.13

Update: There's a great summary of the history of the user agent string at WebAIM: Blog - History of the browser user-agent string.

Roundup: XP performance vs. Vista

2008-08-25T10:53:00.000-07:00

I dug a bit into some anecdotal reports I had heard about XP vs. Vista. It turns out XP performance beats Vista: 32- or 64-bit, service packs or no, in a box or with a fox, in a house or with a mouse.

Benchtest: Windows XP SP3 vs. Windows Vista SP1 | News | TechRadar UK — Jun 23, 2008 — 787MB is an absolutely ludicrous amount for a system to swallow from a fresh boot of a new installation. Windows XP squeezes itself into just one fifth the space.
XP Vista SP1 2008 x86 and x64 benchmark | Vista Home [graph] — Apr 28, 2008
Testing Shows XP Still Outperforms Vista - Software - IT Channel News by CRN and VARBusiness — Apr 04, 2008
exo.blog: Update: Re-Testing Vista w/2GB RAM, Office 2003 — Nov 27, 2007
Windows XP outshines Vista in benchmarking test - CNET News.com — Nov 27, 2007
Benchmark Results - Tom's Hardware : Windows XP vs. Vista: The Benchmark Rundown — Jan 29, 2007

Sure, it's doing more, but is it doing more good? Death match: Windows Vista versus XP | InfoWorld | Analysis | 2008-03-17 | By Randall C. Kennedy

.NET 3.5 sp1 Setup Blocks Itself!

2008-08-14T09:38:00.000-07:00

Um… ooooooooooooo-kay… that seems a bit meta.

I guess I'll click Ignore? Oh look, the progress bar is going backwards. I guess it's a Regress Bar.

Efficacy of .NET StreamReader's detectEncodingFromByteOrderMarks

2008-08-06T11:54:00.000-07:00

The .NET System.IO.StreamReader class has several forms of its constructor that accept a boolean detectEncodingFromByteOrderMarks parameter to look for a byte-order-mark (BOM)/encoding-signature when the file is first read.

When enabled, this feature populates the CurrentEncoding property after the first time the file is read (which can be a simple call to Peek()).

This method only works reliably for encodings that supply a BOM, but since the default encoding is utf-8, several other single-byte encodings are compatible with content in the 7-bit ASCII range.

Here is a sample of how well this feature works with content written in various encodings:

us-ascii: Not detected, but works fine with the default UTF-8 since ASCII is a subset of UTF-8.
utf-7: Not detected, not UTF-8 compatible.
utf-8: Detected correctly. Default encoding anyway.
utf-16/UCS-2: Detected correctly (as utf-16).
utf-32: Detected correctly.
utf-32BE: Detected correctly, but still reads incorrectly in my testing!
unicodeFFFE: Detected correctly.
Windows-1252, iso-8859-1, iso-8859-15, macintosh: Not detected, but shares a significant character overlap with the UTF-8 default (7-bit ASCII).
Various EBCDIC encodings: IBM037, IBM500, IBM870: Not detected, and not read correctly in tests.
UTF-EBCDIC, SCSU, BOCU-1, Punycode, CESU-8, UCS-4*, UTF-1, UTF-9†, UTF-18†: Not supported by the .NET Framework.

For the most part, content using a Unicode encoding of some kind (which include a BOM) have the greatest chance of success, and encodings not listed aren't likely to work. EBCDIC and international encodings, among others, must really be opened using their explicit encoding (meaning they must be anticipated), if they are to be read successfully, which is why you should only produce UTF-8/16/32 content.

* Not recognized as an alias for UTF-32.

† To be fair, these encodings are a joke.

Windows Special Folders Support

2008-07-31T10:48:00.000-07:00

Windows system/special folders, unlike Linux (which has a consistent, reliable filesystem heirarchy), can be in all kinds of crazy custom locations, or use localized text.

As a consequence, you cannot simply use a path to a special folder. You have to look it up first. Nearly all Microsoft APIs/platforms/technologies provide some way to look up these special folders. What's even more complicated is that these lookups seem to be about as varied as the folder names themselves.

Not only do newer technologies reflect completely new folders, as could be expected, but each API seems to provide access to a different subset of special folders, and use different names/keys for the folders, and sometimes provide multiple partial lists via different methods/lookups/constants. Say you want to look up the location of the Windows Recycle Bin, does your API refer to this as CSIDL_BITBUCKET or FOLDERID_RecycleBinFolder, or does it provide any access to it at all?

This reference aims to provide an overview of whether your API supports a given special folder, and where to look for it.

Windows Special Folder Environment Variables

ALLUSERSPROFILE
APPDATA
CommonProgramFiles
CommonProgramFiles(x86)
CommonProgramFilesW6432
ProgramFiles
ProgramFiles(x86)
ProgramFilesW6432
SystemDrive
SystemRoot
TEMP
TMP
USERPROFILE
windir

Windows XP Embedded uses numeric environment variables.

.NET `System.Environment.SpecialFolder` Enumeration

ApplicationData
CommonApplicationData
LocalApplicationData
Cookies
Desktop
Favorites
History
InternetCache
Programs
MyComputer
MyMusic
MyPictures
Recent
SendTo
StartMenu
Startup
System
Templates
DesktopDirectory
Personal
MyDocuments
ProgramFiles
CommonProgramFiles

.NET `System.IO.Path` Methods

GetTempPath()

Windows Scripting Host `WshShell.SpecialFolders` Collection

AllUsersDesktop
AllUsersStartMenu
AllUsersPrograms
AllUsersStartup
Desktop
Favorites
Fonts
MyDocuments
NetHood
PrintHood
Programs
Recent
SendTo
StartMenu
Startup
Templates

Windows Scripting Host `FileSystemObject.GetSpecialFolder()` Method Parameters

SystemFolder
TemporaryFolder
WindowsFolder

Windows CLSID GUIDs

Administrative Tools	`{D20EA4E1-3957-11d2-A40B-0C5020524153}`
Briefcase	`{85BBD920-42A0-1069-A2E4-08002B30309D}`
Control Panel	`{21EC2020-3AEA-1069-A2DD-08002B30309D}`
Fonts	`{D20EA4E1-3957-11d2-A40B-0C5020524152}`
History	`{FF393560-C2A7-11CF-BFF4-444553540000}`
Inbox	`{00020D75-0000-0000-C000-000000000046}`
Microsoft Network	`{00028B00-0000-0000-C000-000000000046}`
My Computer	`{20D04FE0-3AEA-1069-A2D8-08002B30309D}`
My Documents	`{450D8FBA-AD25-11D0-98A8-0800361B1103}`
My Network Places	`{208D2C60-3AEA-1069-A2D7-08002B30309D}`
Network Connections	`{7007ACC7-3202-11D1-AAD2-00805FC1270E}`
Printers and Faxes	`{2227A280-3AEA-1069-A2DE-08002B30309D}`
Recycle Bin	`{645FF040-5081-101B-9F08-00AA002F954E}`
Scanners and Cameras	`{E211B736-43FD-11D1-9EFB-0000F8757FCD}`
Scheduled Tasks	`{D6277990-4C6A-11CF-8D87-00AA0060F5BF}`
Temporary Internet Files	`{7BD29E00-76C1-11CF-9DD0-00A0C9034933}`
Web Folders	`{BDEADF00-C265-11D0-BCED-00A0C90AB50F}`

Win32 `CSIDL` Constants

CSIDL_ADMINTOOLS
CSIDL_ALTSTARTUP
CSIDL_APPDATA
CSIDL_BITBUCKET
CSIDL_CDBURN_AREA
CSIDL_COMMON_ADMINTOOLS
CSIDL_COMMON_ALTSTARTUP
CSIDL_COMMON_APPDATA
CSIDL_COMMON_DESKTOPDIRECTORY
CSIDL_COMMON_DOCUMENTS
CSIDL_COMMON_FAVORITES
CSIDL_COMMON_MUSIC
CSIDL_COMMON_OEM_LINKS
CSIDL_COMMON_PICTURES
CSIDL_COMMON_PROGRAMS
CSIDL_COMMON_STARTMENU
CSIDL_COMMON_STARTUP
CSIDL_COMMON_TEMPLATES
CSIDL_COMMON_VIDEO
CSIDL_COMPUTERSNEARME
CSIDL_CONNECTIONS
CSIDL_CONTROLS
CSIDL_COOKIES
CSIDL_DESKTOP
CSIDL_DESKTOPDIRECTORY
CSIDL_DRIVES
CSIDL_FAVORITES
CSIDL_FONTS
CSIDL_HISTORY
CSIDL_INTERNET
CSIDL_INTERNET_CACHE
CSIDL_LOCAL_APPDATA
CSIDL_MYDOCUMENTS
CSIDL_MYMUSIC
CSIDL_MYPICTURES
CSIDL_MYVIDEO
CSIDL_NETHOOD
CSIDL_NETWORK
CSIDL_PERSONAL
CSIDL_PRINTERS
CSIDL_PRINTHOOD
CSIDL_PROFILE
CSIDL_PROGRAM_FILES
CSIDL_PROGRAM_FILESX86
CSIDL_PROGRAM_FILES_COMMON
CSIDL_PROGRAM_FILES_COMMONX86
CSIDL_PROGRAMS
CSIDL_RECENT
CSIDL_RESOURCES
CSIDL_RESOURCES_LOCALIZED
CSIDL_SENDTO
CSIDL_STARTMENU
CSIDL_STARTUP
CSIDL_SYSTEM
CSIDL_SYSTEMX86
CSIDL_TEMPLATES
CSIDL_WINDOWS

Windows Vista `KNOWNFOLDERID`

FOLDERID_AddNewPrograms
FOLDERID_AdminTools
FOLDERID_AppUpdates
FOLDERID_CDBurning
FOLDERID_ChangeRemovePrograms
FOLDERID_CommonAdminTools
FOLDERID_CommonOEMLinks
FOLDERID_CommonPrograms
FOLDERID_CommonStartMenu
FOLDERID_CommonStartup
FOLDERID_CommonTemplates
FOLDERID_ComputerFolder
FOLDERID_ConflictFolder
FOLDERID_ConnectionsFolder
FOLDERID_Contacts
FOLDERID_ControlPanelFolder
FOLDERID_Cookies
FOLDERID_Desktop
FOLDERID_Documents
FOLDERID_Downloads
FOLDERID_Favorites
FOLDERID_Fonts
FOLDERID_Games
FOLDERID_GameTasks
FOLDERID_History
FOLDERID_InternetCache
FOLDERID_InternetFolder
FOLDERID_Links
FOLDERID_LocalAppData
FOLDERID_LocalAppDataLow
FOLDERID_LocalizedResourcesDir
FOLDERID_Music
FOLDERID_NetHood
FOLDERID_NetworkFolder
FOLDERID_OriginalImages
FOLDERID_PhotoAlbums
FOLDERID_Pictures
FOLDERID_Playlists
FOLDERID_PrintersFolder
FOLDERID_PrintHood
FOLDERID_Profile
FOLDERID_ProgramData
FOLDERID_ProgramFiles
FOLDERID_ProgramFilesX64
FOLDERID_ProgramFilesX86
FOLDERID_ProgramFilesCommon
FOLDERID_ProgramFilesCommonX64
FOLDERID_ProgramFilesCommonX86
FOLDERID_Programs
FOLDERID_Public
FOLDERID_PublicDesktop
FOLDERID_PublicDocuments
FOLDERID_PublicDownloads
FOLDERID_PublicGameTasks
FOLDERID_PublicMusic
FOLDERID_PublicPictures
FOLDERID_PublicVideos
FOLDERID_QuickLaunch
FOLDERID_Recent
FOLDERID_RecordedTV
FOLDERID_RecycleBinFolder
FOLDERID_ResourceDir
FOLDERID_RoamingAppData
FOLDERID_SampleMusic
FOLDERID_SamplePictures
FOLDERID_SamplePlaylists
FOLDERID_SampleVideos
FOLDERID_SavedGames
FOLDERID_SavedSearches
FOLDERID_SEARCH_CSC
FOLDERID_SEARCH_MAPI
FOLDERID_SearchHome
FOLDERID_SendTo
FOLDERID_SidebarDefaultParts
FOLDERID_SidebarParts
FOLDERID_StartMenu
FOLDERID_Startup
FOLDERID_SyncManagerFolder
FOLDERID_SyncResultsFolder
FOLDERID_SyncSetupFolder
FOLDERID_System
FOLDERID_SystemX86
FOLDERID_Templates
FOLDERID_TreeProperties
FOLDERID_UserProfiles
FOLDERID_UsersFiles
FOLDERID_Videos
FOLDERID_Windows

TweakUI Special Folders

CD Burning
Desktop
Document Templates
Favorites
Installation Path
My Documents
My Music
My Pictures
My Video
Programs
Send To
Shared Music
Shared Pictures
Shared Video
Start Menu
Startup

Visual Studio 2008 Setup Project → Add Special Folder

Common Files Folder
Common Files (64-bit) Folder
Fonts Folder
Program Files Folder
Program Files (64-bit) Folder
System Folder
System (64-bit) Folder
User's Application Data Folder
User's Desktop
User's Favorites Folder
User's Personal Data Folder
User's Programs Menu
User's Send To Menu
User's Start Menu
User's Startup Folder
User's Template Folder
Windows Folder
Global Assembly Cache Folder

Visual Studio 2008 Setup Project → File Installation Properties → DefaultLocation [`SpecialFolders`]

AdminToolsFolder
AppDataFolder
CommonAppDataFolder
CommonFiles64Folder
CommonFilesFolder
DesktopFolder
FavoritesFolder
FontsFolder
LocalAppDataFolder
MyPicturesFolder
PersonalFolder
ProgramFiles64Folder
ProgramFilesFolder
ProgramMenuFolder
SendToFolder
StartMenuFolder
StartupFolder
System16Folder
System64Folder
SystemFolder
TempFolder
TemplateFolder
WindowsFolder
WindowsVolume

Way to go, abc.com!

2008-06-25T20:48:00.000-07:00

So my DVR missed the most recent episode of the Mole. I decided, rather than grabbing the episode from a torrent, that I'd try and play their game.

I went to abc.com, and quickly found out that Linux STILL isn't supported (hulu seems to have figured it out).
So, I grit my teeth and boot into my old, neglected Windows partition and go to abc.com again.
I had to install a plugin and restart Firefox.
Finally, after a lengthy download, I chose the episode to watch and got an error.
I noticed that the error is BECAUSE THERE IS NO AD DATA.
Nevermind.

So after all of that, I can't watch the episode because abc.com wasn't able to line up commercials? Brilliant. Advertisers want viewers, so clearly the smartest way to drum up advertising is to lock out viewers!

Firefox 3 Features

2008-06-09T09:40:00.000-07:00

Here's a good demo of the Awesome Bar, plus: find out who Larry is.

Overview of Firefox 3

I also found out this weekend that Firefox 3 has true zoom, like Opera has had for years. If you zoom in, the images and layout all zoom in, not just the text anymore.

How to create your own "smart bookmark" queries in Firefox 3 (saved search bookmarks/history, sort by 20 different properties, filter by domain or term, and limit the number returned):

Create Your Own Smart Bookmarks in Firefox 3

Here's how to restore the original examples:

Restore Firefox 3 Smart Bookmarks

Update: Visual Studio's NIH RegEx Syntax

2008-05-22T10:48:00.001-07:00

It looks like this topic came up a couple of years ago with VS2005 at Coding Horror, including a Microsoft reply.

Hey Jeff (and assorted follow-up posters),
I'm the lead program manager for the team that owns editing and the find/replace dialog in Visual Studio. Our team agrees with your post :)
It is a very oddball regex syntax, and as best we can tell it comes from Visual C++ 2.0. We did want to add additional support for .NET 2.0-style regular expressions in the Visual Studio 2005 release, but unfortunately due to time pressures it didn't make the final list of features. We were able to make a number of bug fixes to the existing engine though, to give some improvement over VS 2003.
We do keep this on our list of things we want to fix. Ideally at some point we'll actually build in a nifty little extensibility point so you can wire up any regex engine you want for searches.
Thanks for the feedback!
Neil Enns
Lead Program Manager
Microsoft Visual Studio
— Neil on July 14, 2006 11:15 AM

So, take heart! A Microsoft team member agreed about the stupidity of the NIH RegEx syntax in 2006, so I'm sure we'll see this fixed any day now.

What we need, as pointed out by another commenter, is another pattern option: Regular regular expressions.

Instead, the closest we currently have is couple of VS add-ons to replace the Find and Replace dialog that support normal regex syntax, an unfamiliar UI, broken English, and VS 2005, and no express editions.

Visual Studio's NIH RegEx Syntax

2008-05-06T08:09:00.000-07:00

Here's a quick phrasebook for Visual Studio's NIH RegEx syntax:

VS Editor RegEx Syntax	Real RegEx Syntax*	Meaning
`{…}`	`(…)`	tagged / captured submatch
`(…)`	`(?:…)`	non-capturing submatch
✗	`(?=…)`	lookahead assertion
`~(…)`	`(?!…)`	negative lookahead assertion / prevent match
✗	`(?<=…)`	lookbehind assertion
✗	`(?<!…)`	negative lookbehind assertion
✗	`(?>…)`	nonbacktracking (greedy) subexpression
`<`	`\<`	start of word
`>`	`\>`	end of word
`\<`	`<`	matches `<` character
`\>`	`>`	matches `>` character
`(<\|>)`	`\b`	word boundary
`~(<\|>)`	`\B`	not a word boundary
✗	`?`	zero-or-one quantifier
✗	`??`	minimal zero-or-one quantifier
`@`	`*?`	minimal zero-or-more quantifier
`\@`	`@`	matches `@` character
`#`	`+?`	minimal one-or-more quantifier
`\#`	`#`	matches `#` character
`^n`	`{n}`	match `n` times quantifier
`\^`	`^`	matches `^` character
✗	`{n,}`	match at least `n` times quantifier
✗	`{m,n}`	match between `m` and `n` times (inclusive) quantifier
✗	`{n,}?`	minimally match at least `n` times quantifier
✗	`{m,n}?`	minimally match between `m` and `n` times (inclusive) quantifier
`\(w,n)`	✗	(replacement expression) left-pad captured group `n` to `w` characters
`\(-w,n)`	✗	(replacement expression) right-pad captured group `n` to `w` characters
`\g`	`\a`	alert / bell
`\h`	`[\b]`	backspace
`\:`	`:`	matches `:` character
`:i`	`([a-zA-Z_$][a-zA-Z0-9_$]*)`	identifier
`:q`	`(("[^"]")\|('[^']'))`	quoted string
`:h`	`([0-9A-Fa-f]+)`	hexadecimal number (not including any prefix, e.g. `0x` or `\x` or `\u`)
`:n`	`((\d+.\d)\|(\d.\d+)\|(\d+))`	rational number
`:w`	`(\p{L}+)`	letters
`:b`	`[ \t]`	space or tab (like `\s` without `\n` or `\v`)
`:z`	`\d+`	integer (one or more decimal digits)
`:a`	`\w`	word / alphanumeric character
`[^:a]`	`\W`	non-word / non-alphanumeric character
`:c`	`\p{L}`	letter character (like `\w` without the _)
`:d`	`\d`	decimal digit
`[^:d]`	`\D`	non-decimal-digit character
`:U`	`\p{U}`	matches Unicode character category `U`
✗	`\p{IsBlock}`	matches characters in Unicode named block `Block`
`[^:U]`	`\P{U}`	does not match Unicode character category `U`
✗	`\P{IsBlock}`	does not match characters in Unicode named block `Block`
`:Al`	`\p{L}`	letter
`:Nu`	`\d`	decimal digit
`:Pu`	`\p{P}`	punctuation character
`:Wh`	`\s`	whitespace character
`[^:Wh]`	`\S`	non-whitespace character
`:Bi`	?	bidirectional character
`:Ha`	`\p{IsHangulJamo}`	Korean Hangul and combining Jamos
`:Hi`	`\p{IsHiragana}`	hiragana character
`:Ka`	`\p{IsKatakana}`	katakana character
`:Id`	?	ideographic characters, such as Han and kanji

*This is the syntax supported by everything else, including the .NET System.Text.RegularExpressions library. For some reason, Microsoft decided to create a new syntax just for the Visual Studio editor! But they just don't have the resources to implement "your favorite standard".

Anything that's not on the list should be the same for NIH Regex patterns.

Dear Microsoft, We Need XSLT 2.0

2008-04-22T08:16:00.000-07:00

XSLT2 Is Useful

Listen, I really appreciate the work Microsoft has done on IE8 standards, and I think LINQ to XML really improves on the W3C XML DOM, but I also need to produce XML from templates.

You see, ASP.NET really only produces HTML from templates. If you want to produce RSS, Atom, POX, SVG, P3P, XSLT, or XSL-FO from templates, probably the best way to do this (for flexibility, maintainability, and readability) is with XSLT Simplified Syntax.

Example

<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom"
      xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
      xmlns:fn="http://www.w3.org/2006/xpath-functions"
      xsl:exclude-result-prefixes="xsl fn"
      xsl:version="2.0">
  <title>Press Releases</title>
  <link href="http://example.org/"/>
  <updated>
    <xsl:value-of select="fn:max(//story/@releasedate)"/>
  </updated>
  <id>urn:uuid:7b2845b6-cc0d-4a64-92a4-2c0f33a398a2</id>
  <xsl:for-each select="//story">
    <entry>
      <title>
        <xsl:value-of select="@name"/>
      </title>
      <link href="http://example.org/"/>
      <id>
        <xsl:text>http://example.org/press/</xsl:text>
        <xsl:value-of select="fn:format-dateTime(@releasedate,'YYYYMMDDTHHmmss')"/>
      </id>
      <updated>
        <xsl:value-of select="@releasedate"/>
      </updated>
      <content type="xhtml" xml:lang="en">
        <xsl:copy-of select="div"/>
      </content>
    </entry>
  </xsl:for-each>
</feed>

This is more readable and maintainable than using the DOM to build up the document, or a full XSLT stylesheet. It is much more approachable, since it is a simple template, which allows trivial changes without having to understand the complete XSLT syntax. Plus, a minimal .NET HTTP handler can apply XSLT to source XML (perhaps produced by SQL Server) without much code to maintain.

The Problem

Anyone that's tried to use XSLT 1.0 quickly finds that they can't use most of the modern methods they've become used to in other languages: regular expressions, date/time/number parsing and formatting, grouping, data structures, Unicode support, date/time/timezone/duration data types and functions, set manipulation, aggregate functions, and converting relative URLs to absolute ones, to name some common examples of things that are simply prohibitively difficult and complex in XSLT 1.0 (particularly using the simplified syntax).

Of course these have all been addressed in XSLT 2.0 and XPath 2.0, released as W3C Recommendations over a year ago, moving XSLT from curiosity to practical templating engine.

The .NET XSLT library still only supports XSLT 1.0 and XPath 1.0.

The emphasized functions in the above example give an indication of why this is a problem. How can you provide aggregate functions and date/time formatting without those functions without losing the benefits of the XSLT Simplified Syntax? There are really only two practical answers.

Saxon

Buy Saxon (or use Saxon Basic, which is free).

Benefits

You support Michael Kay, XSLT ninja.
The market for XSLT grows.
You won't have to make changes to all of your templates when Microsoft finally supports XSLT2.
You spend no time trying to figure out how to work around the lack of a simple XSLT2 feature.
You get used to XSLT2.

Kludgy Workarounds

If your needs are more modest, you could try converting your stylesheet to XSLT1 and work around the limitations of that implementation.

Benefits

You don't spend time learning a third party library.
You don't have to manage a third party library when deploying your system.
You spend no money, provided you do not assign any value to the time required for you to come up with workarounds.

Obviously, you only want to use this approach if you have to, or if there are only a few simple workarounds to get what you want into the template, such as using the SQL Server CONVERT function to format date/time/currency/numeric values before building the XML document, or altering the XML DOM to add or substitute formatted or aggregate values before applying the stylesheet.

Conclusion

Microsoft really needs to focus some resources on getting XSLT 2.0 into the next release of .NET. Clearly, between IE8 and OOXML, standards support is on Microsoft's TODO list, and there's no question about their committment to XML. Microsoft even knows that their customers want XSLT2. The only question that remains is when market pressure will intensify the need to implement this omission. Be sure to contact Microsoft's XML Team to request a timeline for XSLT2.

What happens when people don't know what they're doing: DoNotReply.com and RFC 2606

2008-03-26T08:48:00.000-07:00

People without a basic understanding of RFC 2606 have been using a From or Reply-To address of donotreply@donotreply.com, and surprise!, that domain exists.

This implies not just job incompetence, but a manner so cavalier and naïve that the person responsible didn't even consider typing donotreply.com into a browser.

This is a level of technical malpractice worthy of termination, IMO.

Remember, the people that designed the Internet were generally pretty clever folks. If you need a phony domain name (akin to a 555 phone number), for example, there's a pretty good chance someone has set aside a few of those.

.test
.example
.invalid
.localhost
example.com
example.net
example.org

Note: IANA has even set aside an example MIME supertype and subtypes for all of the other supertypes.

Re: The Internet Explorer 8 User-Agent String

2008-03-25T15:46:00.000-07:00

Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0)

Microsoft, why are you still lying about being Mozilla?

Seriously, it's just wasted network traffic and drive space and extra parsing for log analyzers.

I see many commentors, in the recent IEBlog post about the IE8 user agent string, requesting an end to the "Mozilla/4.0 (compatible; " nonsense:

I see only two people defending the 1990s style user agent string:

This seems to indicate a pretty clear preference from the developer community.

There will be a few sites that incorrectly slam the door on their guests simply due to an unknown UA string, but they clearly need a wake-up call anyway. Whitelisting user agent strings is stupid, and needs to be stopped now.

P.S.: Thanks for making standards mode the default in IE8, Microsoft! I'm not kidding or being sarcastic. You've really improved the lives of developers in the long run, whether or not everyone understands it now.

Vista at MIX08

2008-03-10T10:25:00.000-07:00

At nearly every session, Windows Vista froze briefly and flickered several times in a way that interrupted the session presentations significantly. It was obvious that the presenters were afraid the machine had just crashed, though the desktop eventually returned. Usually, this was at a time when the input to the projector was changed, and I'm guessing that changed the video subsystem enough to cause Vista Content Protection to restart the video subsystem.

This seems to confirm: Peter Gutmann was right.

MIX08 News

2008-03-06T09:54:00.000-08:00

IE8 will finally support data: URLs!
SSDS provides REST, SOAP, or APP access to a SQL data store.
There have been many changes to ASP.NET Dynamic Data controls since they were introduced at the last MIX: everything seems more template-based, uses attribute decorating in addition to SQL schema info to automatically build GridView and DetailsView/FormView controls.
Microsoft is REALLY into REST.
Silverlight can be worked into any conversation.
IIS7 is pretty exciting, particularly for developers with hosted applications (or corporations in a similar situation), since nearly every IIS setting is now in the web.config file.
Hosts in IIS7 can delegate features with impressive granularity, or lock them down.
The IIS7 admin console can connect through port 80.
IE8 is pushing something called WebSlices.

MIX08: Day One

2008-03-05T09:08:00.000-08:00

This year, Microsoft handed out purses to all attendees, made out of the finest white tarpaulin. To complete the look, white (or green or black) berets were available, for anyone that wanted to rock the Jamie Hyneman look. No matching shoes yet, though.

No expense has been spared, as Microsoft has made a full 11Mb/s 802.11b Internet connection available to anyone directly within the conference rooms. Internet access from within the hotel rooms can be had for only $10/day.

Last night, we attended a Microsoft screening of The King of Kong (2007), and tonight the subject of the movie will try to break the current Donkey Kong high score.

Warning: Do not buy Microsoft input devices

2008-03-01T10:57:00.000-08:00

My Microsoft Sidewinder Strategic Commander doesn't work under Vista (64-bit). It's pretty clear that Microsoft considers even a USB input device to have an expiration date, since they appear to have no plans to add this support.

I bought Bioshock recently, and it also seems obvious that Microsoft paid the developer to make using any game controller other than the XBOX 360 controller virtually impossible to set up.

So, I guess in a few years, when the next version of Windows is released and doesn't support even that controller, I'll be stuck with keyboard and mouse.

Though this approach works really well for Microsoft, and stands as the familiar degenerate case of Capitalism, I'm tired of spending money unnecessarily. Is this what a stand-alone hardware company or OS vendor would do? This isn't anticompetitive?

Windows Error du jour

2008-02-21T08:11:00.000-08:00

What is the sound of one pointer hiding?

Innovative Input: AlphaGrip AG-5 3D Keyboard

2008-02-05T19:47:00.000-08:00

AlphaGrip, a Handheld USB Computer Keyboard and Trackball for Anywhere Computing

Apparently, you can get a AlphaGrip AG-5 3D Keyboard and Trackball, USB at TrackballWorld.com

I'm still waiting for my Panther XL successor, though.

Election '08: Meat Update

2008-01-16T08:07:00.000-08:00

In case you only suspected candidates are just meat products...

70-551 UPGRADE: MCAD Skills to MCPD Web Developer by Usingthe Microsoft® .NET Framework

2008-01-03T16:05:00.001-08:00

70-551 is harrowing.

Nowhere that I've been able to find does it mention that you are simply taking 70-528, 70-536, and 70-547 back-to-back.

In fact, the only place that gives any clue to the ridiculous, extreme length of this testing trilogy is CramSession, which claims the test is "50 questions" and "90 minutes", which is completely wrong.

I was completely blindsided when I walked in to take a test in November, and had to sit for HOURS AND HOURS taking THREE &$@^$$#^* TESTS IN A ROW!

HEY, MICROSOFT

D'ya think maybe you could at least give your victims a clue as to the scope, or at least the ORDER OF MAGNITUDE of your damn tests?

Dear Microsoft, we need an updater

2008-01-02T10:37:00.000-08:00

Update 4: add CNET TechTracker [via Lifehacker, via Digital Inspiration].

Update 3: add RadarSync [via Lifehacker, via MakeUseOf].

Update 2: add SUMo (Software Update Monitor) (thanks to a comment by MaribelPelops on the same Lifehacker article).

Update: and add Update Notifier to the mix (thanks Lifehacker).

Every time I install a program, icons are left all over my Start menu and desktop, mostly crap I don't need. Also, enough with the EULAs already!

Most of all, I'm tired of some programs staying in memory (Java, WinAmp Agent, WinZip, etc.) to check for updates periodically, some programs adding separate Scheduled Tasks (Apple stuff, etc.), some programs checking at startup (often slowly, in the main thread: Adobe Reader, etc.), and most programs requiring I manually keep stuff up to date by watching their stupid website and re-installing when there is an update. I'm lucky if the auto-update checks even have the capacity to install the update, rather than just throwing the download web page at me to do myself. I'm also lucky if I don't have to uninstall the previous version, even this many years after the introduction of the MSI. Of course, I also have to constantly monitor vulnerability sites like US-CERT: United States Computer Emergency Readiness Team, since commercial entities can't be trusted to be honest about unpatched vulnerabilities.

Oh, there are many attempts to bring some measure of automation to installation and updates on Windows, but despite AppUpdater, win-get, app-get, UpdateStar, AppSnap, filehippo.com Update Checker, Secunia Personal Software Inspector, and others, most of them fall pathetically short.

To summarize: I spend all my time in Windows installing, monitoring for updates, updating, and cleaning up afterwards, and it SUCKS.

Stop fighting with me about who really owns my machine (premium content DRM), putting lipstick on the pig (Aero), crippling network speed when music is played back (Multimedia Class Scheduler), implementing half-assed web standards (IE7), asking users what components of an application to install (MSI), and fix the real problem with Windows: installs and updates.

Sincerely,

Brian

List of lists of 2007 lists

2008-01-01T14:08:00.000-08:00

In the spirit of meta-listing, I give you the First Annual List of Lists of 2007 Lists (including other lists of lists of 2007 lists), AKA the Top of the Tops of the Tops:

Why OPML?

2007-12-19T16:22:00.000-08:00

So, when exporting or importing a list of feeds, why are we using OPML, when any feed reader already has an Atom and/or RSS parser? It seems like those formats would work as well for a list of links.

configSource: "It must refer to a file in the same directory or in a subdirectory as the configuration file."

2007-12-05T15:11:00.000-08:00

Since the configSource attribute, to include .NET config file sections from external files, only allows relative paths and no parent paths, it severely limits the ability to centralize configuration settings for multiple config files.

Unless you use NTFS hard links or junction points, of course.

For example, if you have a ConnectionStrings.config file that you want to reference from your .NET 2.0 machine.config and your .NET 3.5 machine.config, you can create a directory within the .NET 2.0 CONFIG directory called ConnectionStrings, then create a hard link (for files, using fsutil hardlink) or junction point (for an entiry directory, using junction) with the same name within the .NET 3.5 CONFIG directory. When you put the ConnectionStrings.config in either directory, it will exist in both, so you can point to ConnectionStrings\ConnectionStrings.config with a configSource attribute in both machine.config files, and still only have a single file to maintain.

Telco Dereliction

2007-10-26T16:03:00.000-07:00

All of the US telcos have really let us down with lousy bandwidth and terrible upload speeds, while taking $200 billion dollars from taxpayers.

Globbing (using filename wildcard patterns) in C#

2007-09-26T09:02:00.000-07:00

The Directory.GetFiles() method in the .NET CLR is unsatisfying for anyone that just wants to use simple filename wildcard patterns. Before you can even call GetFiles(), there are several steps you have to perform manually. Minimally, you have to split the directory from the filename pattern, then you have to convert the directory to an absolute path. If you want to process an array of delimited pattern lists, there are even more steps. Here's a simple recipe:

public static class Glob
{
    static public string[] GetFiles(string[] patterns)
    {
        List<string> filelist= new List<string>();
        foreach(string pattern in patterns)
            filelist.AddRange(GetFiles(pattern));
        string[] files= new string[filelist.Count];
        filelist.CopyTo(files,0);
        return files;
    }
    
    static public string[] GetFiles(string patternlist)
    {
        List<string> filelist= new List<string>();
        foreach(string pattern in 
            patternlist.Split(Path.GetInvalidPathChars()))
        {
            string dir= Path.GetDirectoryName(pattern);
            if(String.IsNullOrEmpty(dir)) dir= 
                Directory.GetCurrentDirectory();
            filelist.AddRange(Directory.GetFiles(
                Path.GetFullPath(dir),
                Path.GetFileName(pattern)));
        }
        string [] files= new string[filelist.Count];
        filelist.CopyTo(files,0);
        return files;
    }
}

This only resolves patterns in filenames, not paths, but may serve the needs for this common task for most uses, and provides a starting point for something more comprehensive.

Media Queries: Opera and Safari

2007-06-13T09:11:00.000-07:00

For about a year now, both Opera and Safari have supported CSS3 Media Queries, which allow separate CSS rules based on screen or window (viewport) size, color/grayscale depth, aspect ratio, resolution, tv progressive/interlaced, or grid devices (like dumb terminals or other tty-style devices).

This allows a number of capabilities:

More layout columns for the growing audience of 30-inch monitors and widescreen laptops.
Single-column layout when only a small or narrow view of the page is available (maybe due to a handheld device, a cheap video card, or an older computer).
Less graphically intense layout for "grid" devices (such as telnet sessions or e-book devices).
Different backgrounds for browsers with serious color limitations, or black and white displays (maybe for color-blind or low-vision users).
Different layout for pivoted monitors or other displays that are taller than they are wide.

Firefox and Internet Explorer need to get on board before this is widely useful, but now that Safari is available for Windows, it may add just a little bit more pressure.

Here's how to Team Build a Web Application Project

2007-05-17T17:29:00.000-07:00

Update: Aaron Hallberg : Team Build and Web Deployment Projects

First, create a Web Deployment Project for your Web Application Project. (If you have a Web Site Project, you will need to convert the Web Site Project to a Web Application Project before continuing.)
Set the Output Folder to the destination UNC for the Debug and Release build types.
Set any other WDP properties appropriate to your environment, such as Web.config file section replacement.
Make sure your build server has VisualStudio and VS SP1 installed. If the server does not have the C:\Program Files\MSBuild\Microsoft\WebDeployment\ folder, you may need to install Web Deployment Projects as well.
Create fixup.TeamWebDeployment.targets (see below) in C:\Program Files\MSBuild\. Normally, the Team Build drop location overrides the Output Folder, so this file adds a second compile task to put the web app where you wanted it to go in the first place.
Add an import directive for $(MSBuildExtensionsPath)\fixup.TeamWebDeployment.targets.
Create the team build, choosing your solution and build type, but you MUST choose Mixed Platforms as the Platform, or things will go very badly for you.
Cross your fingers and run the build.

fixup.TeamWebDeployment.targets

<Project DefaultTargets="Build" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
 <Target Name="AfterBuild" Condition="$(OutputPath) != $(WDTargetDir)">
  <AspNetCompiler
   PhysicalPath="$(_AspNetCompilerSourceWebPath)"
   TargetPath="$(OutputPath)"
   VirtualPath="$(_AspNetCompilerVirtualPath)"
   Force="$(_Force)"
   Debug="$(DebugSymbols)"
   Updateable="$(EnableUpdateable)"
   KeyFile="$(_FullKeyFile)"
   KeyContainer="$(_AspNetCompilerKeyContainer)"
   DelaySign="$(DelaySign)"
   AllowPartiallyTrustedCallers="$(AllowPartiallyTrustedCallers)"
   FixedNames="$(_AspNetCompilerFixedNames)"
   Clean="$(Clean)"
   MetabasePath="$(_AspNetCompilerMetabasePath)"
  />
  <RemoveDir Condition="'$(DeleteAppDataFolder)' == 'true'" Directories="$(OutputPath)\App_Data"/>
 </Target>
</Project>

Also, I have to ask any MS folks out there: Why C:\Program Files\MSBuild\ ? Why not C:\Program Files\Common Files\MSBuild\ or C:\Documents and Settings\<current/all user(s)>\Application Settings\Microsoft\MSBuild\ ? This adds yet another customizable location that needs to be backed up.

VS TFS Build: Another day completely wasted

2007-05-14T16:43:00.000-07:00

I've been trying to set up a Team Build for a simple web application for about a week now, and I just don't see the point anymore.

Sure, it integrates into Visual Studio, but that represents ROI for Microsoft, not me.

I've come from scheduling build scripts (.cmd -fu), with some Ant/nAnt experience to the nightmare world of the overdesigned MSBuild, which is just completely opaque. With a syntax that makes a simple tag reference impossible (properties are assigned by the tag name), too much confounding magic (this property or item or target has a special name that is used somewhere deep within the Web Deployment target file, and good luck finding a simple reference for that), and a completely impenetrable variable syntax ( @(FilesToCopy –> '$(MSBuildProjectDirectory)\Bin\%(Filename)%(Extension)') ? Are you kidding me?), and the ever-human-readable GUIDs (also impossible to find a quickref for) sprinkled throughout.

I've found exactly one book, published last year (not even by MS Press), on MSBuild. Other books may provide an appendix or something, but I haven't found anything helpful.

Today I finally got my Web Deployment Project to build successfully as a Team Build, but nothing showed up in the WDP Output Directory. Great, so it isn't working, and there is no error message to tell me why. WTF?

Side topic: What are the intended purposes of Web Sites, Web Application Projects, and Web Deployment Projects? How do they compare? Why do all three exist? Am I using the wrong thing?

Why am I doing this? Because it's the Microsoft direction? What do I gain? As far as I can tell, the primary differences are some better reporting and auditing, integration with the VS Team Process Template (once we decide to try customizing that again), and that our QA staff will now need to open VS to start a test build, where they could just open \\<server>\admin$\Tasks before.

In the meantime, I am hemorrhaging hours, and seriously wondering...why?

MIX07: Embarrasing Moments for Microsoft

2007-04-30T09:21:00.000-07:00

Update: Another really good one.

When we arrived at the Las Vegas airport, one of the baggage carousels had a monitor displaying a Windows error report.

When we got to the hotel, we passed a monitor with a BSOD.

This morning, on the way to the keynote, we passed several monitors that indicated each room's schedule. One had a BSOD, another was flickering the old "You have unused icons on your desktop" tooltray balloon.

A simple copyright question

2007-04-16T10:37:00.000-07:00

I'm probably missing something obvious, but can someone explain to me why a still image recording of someone is the property of the person performing the recording (the photographer), and an audio recording of someone is the property of the subject of the recording (the musician)?

Logitech 3Dconnexion = Cyberman2 ?

2007-04-09T07:36:00.000-07:00

I am always excited by new input devices, so I'm intrigued by the Logitech 3Dconnexion, though I have to say it bears more than a passing resemblance to the disappointing Cyberman2 [2].

Easy XML serialization of complex objects with C#

2007-04-03T09:08:00.000-07:00

If you've got complex, XML-serializable objects you'd like to trace easily, such as SOAP objects, here's a handy way to dump their contents to an XML string:


using System.IO;
using System.Text;
using System.Xml;
using System.Xml.Serialization;

public class XmlExporter
{
 /// <summary>
 /// Serialize an object as an XML string.
 /// </summary>
 /// <param name="o">The object to serialize.</param>
 /// <returns>The object, serialized as an XML string.</returns>
 public static string ToString(object o)
 {
  if (o == null) return "<NULL />";
  string str;
  Encoding utf8 = new UTF8Encoding();
  XmlTypeAttribute xmlatt = (XmlTypeAttribute)Attribute.
   GetCustomAttribute(o.GetType(), typeof(XmlTypeAttribute));
  using (MemoryStream ms = new MemoryStream())
  {
   XmlTextWriter x = new XmlTextWriter(ms, utf8);
   x.Formatting = Formatting.Indented;
   XmlSerializer xs = xmlatt == null ? new XmlSerializer(o.GetType()) : 
    new XmlSerializer(o.GetType(), xmlatt.Namespace);
   xs.Serialize(x, o);
   str = utf8.GetString(ms.GetBuffer(), 0, (int)ms.Length);
   x.Close();
   ms.Close();
  }
  return str;
 }
}

Is Microsoft run by dilettantes?

2007-04-02T07:55:00.000-07:00

The standard .NET class library cannot natively:

work with shorcut (.lnk) files
modify configuration settings
parse command-line options
enumerate ODBC DSNs
determine the data type of a stream by magic number

MythTV config breakthroughs

2007-04-01T18:27:00.000-07:00

At Re: Help Fixing overscan on HDTV, I finally found precisely what I needed to get the correct 1366×768 resolution to my Westinghouse W4207:

williammanda
Re: Help Fixing overscan on HDTV
FWIW
I have found to get the native resolution for my monitor is to have it probed by nvidia. This setup works best so far across many debian setups I have tried. I have a westinghouse 42" lcd native 1366 x 768 and the best I can get is 1360 x 765. I will use the prior posted xorg.conf file as an example
Section "Monitor"
Identifier "SONY TV"
#HorizSync 15.0 - 46.0
#VertRefresh 59.0 - 61.0
#Option "DPMS"
#Option "IgnoreEDID"
Option "UseEDID" "True"
#DisplaySize 720 405
#ModeLine "1280x720@60" 74.48 1280 1336 1472 1664 720 721 724 746 -hsync +vsync
#ModeLine "1216x648@60" 74.48 1216 1272 1400 1664 684 685 688 746 -hsync +vsync
EndSection

Section "Device"
Identifier "NVIDIA Corporation NV18 [GeForce4 MX 440 AGP 8x]"
Driver "nvidia"
EndSection

Section "Screen"
Identifier "Default Screen"
Device "NVIDIA Corporation NV18 [GeForce4 MX 440 AGP 8x]"
Monitor "SONY TV"
#Option "FlatPanelProperties" "scaling = centered"
DefaultDepth 24
SubSection "Display"
Depth 24
Modes "nvidia-auto-select"
EndSubSection
EndSection
Let me know how it works out.
Thanks

Also, I successfully put my network script in /etc/init.d/bootmisc.sh:


#!/bin/sh
iwconfig ath0 essid Springfield rate auto
ifconfig ath0 down
ifconfig ath0 up
dhclient ath0
/etc/init.d/ntp stop
/etc/init.d/ntp start