The short-form week of 14–20 Dec 2009

21 Dec 2009; last modified 13 Mar 2011

The week in review, 140 characters at a time. This week, 31 messages in 35 conversations. (With 2 favorites.)

This document was created automatically from my archive of my Twitter stream. Due to limitations in the Twitter API and occasional glitches in my archiving system, it may not be 100% complete.

In a conversation that started on Monday at 02:12am

Maybe essentially what I want is a transformation language for RDF. (One that can create things other than RDF.)—@JeniT
@JeniT re RDF xform: @ndw and Damain experimented with DOM interface on to RDF graph (not syntax) thence xslt on RDF. treehugger and rdftwig—@skwlilac
@skwlilac I remember @ndw's work. Seems to me that XSLT 2.0 could work over RDF/XML reasonably easily with user-defined functions.—@JeniT
@JeniT @skwlilac It's probably worth exploring with XSLT 2.0. At the time, I was limited to 1.0.—@ndw

In a conversation that started on Tuesday at 10:14am

@JeniT You're lucky, I once felled a tree that hit my house - fortunately only a glancing blow.—@hadleynet
@hadleynet That's why you tie ropes to them *before* you chop them down!—@ndw

Tuesday at 12:08pm

@ndw Its a learning process. My major takeaway was to employ professionals when trees are within reach of the house.—@hadleynet

Tuesday at 03:10pm

I want the country my forefathers died for, not the country my current leaders kill for.—@merlyn

Tuesday at 05:15pm

Turning my eyes away from twitter: folks I follow are discussing the finale of #dexter which I haven't seen yet.—@ndw

In a conversation that started on Tuesday at 09:01pm

First 47 people of the "let me know when GET LAMP is ready" list now notified. Now for the huger one on my old FreeBSD box.—@textfiles
@textfiles Dunno which group I was in, but "ordered".—@ndw

Tuesday at 10:10pm

Holy crap! #dexter—@ndw

In a conversation that started on Wednesday at 05:01am

Oh the meeting was supposed to start at 10:*30*! No wonder no one's here.—@JeniT
@JeniT I once drove 80 miles to a meeting to discover I had arrived a month early.—@michaelhkay
@michaelhkay So did you wait around, or did you drive home? :-P—@ndw

In a conversation that started on Wednesday at 05:04am

CALS tables and XSLT - its going to be one of those days ...—@al3xbrown
@al3xbrown The DocBook XSL stylesheets have complete coverage of CALS tables, I believe.—@ndw
@ndw yup :-)—@al3xbrown

Wednesday at 03:36pm

I think this is a RT of @stephenfry, but I'm not sure. A*maz*ing in any event.—@ndw

Thursday at 07:28am

♺@MarkLogic Register now for the Mark Logic User Conference 2010 to get the early bird special! #MLUC10—@ndw

Thursday at 10:59am

Because not encrypting video feeds from drones in a war zone seemed like such a good idea. (via @elharo)—@ndw

In a conversation that started on Thursday at 12:16pm

Facebook Revenues Screaming Toward $1 Billion. I've never clicked a Facebook ad. I wonder who does?—@ramblingman
@ramblingman Certainly, but wouldn't you say the same about Google?—@avernet
@avernet indeed. once in a rare while I actually do click a Google ad, but thus far never a Facebook.—@ramblingman
@ramblingman Possible reason: tech-savvy folks click on ads less often ( What'll happen as folks get more tech-savvy?—@avernet
@avernet @ramblingman I think tech-savvy folks run adblockplus and flashblock and never even *see* the ads. Or is that just me?—@ndw

Thursday at 02:07pm

You must provide a new password. Not more than 10 chars. No doubled characters. Must not start with a digit. ... #FAIL—@ndw

Thursday at 04:11pm

In Perl: =f(); =a; eq is true. g() fails, g() succeeds. What can be different about two strings that are eq?—@ndw

In a conversation that started on Thursday at 04:11pm

In Perl: $x=f(); $y=a; $x eq $y is true. g($x) fails, g($y) succeeds. What can be different about two strings that are eq?—@ndw
@ndw Tainting? Dualvars? It would be clearer with a bit more code.—@robinberjon
@ndw [What can be different about two strings that are eq?] cf XPath: $x=$y, $x instance of xs:Name = true, $y instance of xs:Name = false.—@michaelhkay
@michaelhkay Yeah, but pass either of those to a function expecting an xs:string argument and they behave the same, right?—@ndw
@ndw [eq strings behave the same, right?] Not if the function does an "instance of" test.—@michaelhkay
@michaelhkay [eq strings behave the same, right?] Also of course collations can make ("a" eq "b") true if you want.—@michaelhkay
@ndw not necessarily, depending on what f() does.—@martin_probst

In a conversation that started on Thursday at 04:17pm

what? you're using bareword a? so $x is 'a'. and what is g() ? you mean f(). Anyway, print 1 if undef eq q{} ..... re:—@terrisCA
@terrisCA No, the quotes got lost. $x and $y are strings that are eq. Deep in the bowls of g() things go awry but $x and $y are *the same*—@ndw

Thursday at 04:29pm

@ndw overloading perhaps?—@perigrin

In a conversation that started on Thursday at 04:29pm

@ndw also undef eq ''—@perigrin
@perigrin No, the strings are both fairly long path names. /path/to/my/file.pdf—@ndw

Thursday at 04:41pm

For total, brain-melting weirdness, print MIME::Base64::encode() *fixes* so that it works. WTF!?—@ndw

Thursday at 04:43pm

@ndw the people on suggest something *really* funky is going on then, and to try Devel::Peek—@perigrin

Thursday at 04:44pm

print MIME::Base64::encode($x), I meant—@ndw

Thursday at 04:52pm

@perigrin One reports PV=..."x"\0 [UTF8 "x"], the other isn't UTF8. That must be the problem, now how to unravel...—@ndw

Thursday at 04:56pm

use Encode; $x = decode("utf8", $x) removes the magic but appears to leave it in utf8. And it still doesn't work.—@ndw

In a conversation that started on Thursday at 05:03pm

@ndw some of us grew out of perl a decade ago and discovered real man languages—@DALDEI
@DALDEI *snort*—@ndw

Thursday at 07:08pm

♺@danja must be repeated, so cool: (via @danbri)—@ndw

In a conversation that started on Thursday at 07:15pm

My mobile phone has a goddamn Unix shell, an integrated Jabber client and uses apt-get to update it's software. So awesome.—@RussB
@RussB That does look like a sweet phone. And I'm starting to think about—@ndw

Thursday at 07:23pm

hueyPRO Calibration sucks. Calibrated 2 days ago, got 14 day warning this morning. Recalibrated. Just got 14 day warning again.—@ndw

Friday at 05:46am

In Java, general problem with impl equals() properly. RT @michaelhkay @ndw [eq strings behave the same] Not if fn does "instance of" test.—@vojtechtoman

Friday at 06:28am

My most influential followers are : @ndw, @martin_probst, @jpcs, @gazdovsky, @tony_vipros. Go to to see yours.—@adamretter

Friday at 08:38am

Refactoring error handling in XML Calabash. Oh, joy.—@ndw

In a conversation that started on Friday at 11:31am

I don't normally check my junk mail these days. But I did today, and found an RFQ for 8 Saxon licenses. Email is broken, how do we fix it?—@michaelhkay
@michaelhkay Not using email, instead asking folks to fill out an online form; adding a captcha if that form is getting spammed.—@avernet
@avernet Please, no. I want to be productive when I'm not online as well as when I am. The answer is charging for mail, but I don't know how—@ndw
@ndw Don't get me wrong: I am not arguing we replace email by web forms. It is popular nowadays to hate email; I am not in that camp.—@avernet
@ndw But when you got lots of orders coming in, web forms are a must. We don't buy books from Amazon by emailing, do we?—@avernet
@michaelhkay Maybe its time to move over to an order-entry system instead of emails for orders. I can suggest one (personal plug)—@DALDEI
@DALDEI There will always be emails asking "can you tell me which product I need?". But perhaps a web-based contact form would help.—@michaelhkay
@michaelhkay A web based contact form is a very easy and useful tool. You can whitelist the "From" so you dont miss emails. Thats critical—@DALDEI

Friday at 04:29pm

♺@fiberartisan RT @remarkablogger: Amazing video on Google Chrome - No, really: it's Amazing!—@ndw

Saturday at 01:36am

best walking-octopus-carrying-a-coconut-shell video I've seen all year (and it's December)!—@shepazu

In a conversation that started on Sunday at 05:00pm

Trying to unify XProc errors and step errors into a single listener for apps like oXygen. I wonder if I'm going about this the wrong way.—@ndw
@ndw Probably not, but why do you think it might be wrong?—@abcoates