weblog » archives
2002 09 29
These people have coded a spellchecker add-on for phpBB2 called phpSpell, and one thing that stroke me is that there’s only one .dic file that one can use with it. And it’s not in the OpenOffice.org format because their dictionary installer doesn’t support the OOorg word modifiers (such as /Y to say the word can be suffixed by -ly).
So I’m taking up the task of writing an installer that supports OOorg modifiers, just for the challenge it represents, and for the great thing it will be to let users be able use the phpSpell script with OOorg dictionaries (I for one, am planning to incorporate phpSpell in b2, so that’s a great motivation to code this support !).
Talking about words, here’s a Googlewhack: hayfield interlocutrice.
2002 09 28
2002 09 27
This is our pinball. It’s a very hard one, it won’t let you score much before it randomly makes you lose the ball in a weird twist of fate.
It’s got a great special feature though: MULLETS. Read the rest of this entry »
My mother and I have tried everything, from house-cleaning products to rotten meat, and the bastard is still alive.
If you have any recipe for poisoning a dog, please contact me.
I am serious.
2002 09 25
All that’s required is the addition of an ID=”yoursite” attribute to the BODY element of each of your site’s pages.
Wouter Demuynck talks about this CSS tip in these terms:
Doesn’t hurt anyone, might be useful to someone. So, why not?
All the reason to use it ! Now if you have a cool modification of tidakada.com, please send me a screenshot and the custom style, and me will love you long time.
Found the link to ESF at Simon’s, couldn’t resist drawing a parallel between ESF and my joke-proposal of RSS 7.0, the difference being ESF is tab-separated values while RSS 7.0 would have been comma-separated values. And I read ESF is actually serious.
…Now, I can scream my lungs out: “You finally did it. You maniacs. You blew it up. God damn you ! God damn you all to hell !“
I find it a bit disappointing that Hixie’s line about Pingback being more transparent than TrackBack was mainly the only thing that Ben Trott commented on.
And that Mark Pilgrim linked to Hixie’s announcement of Pingback 1.0 instead of linking to the specification itself.
After RSS 1.0 versus 2.0, must we see a debate “TrackBack vs Pingback” ?
One thing certain in my opinion, is that auto-discovery from embedded RDF metadata as proposed by Ben is more of a temporary hack than a durable solution: if the next day MT changes its auto-discovery method, making embedded RDF metadata useless, people using new versions of MT would stop embedding RDF and people using the older version of MT that needs embedded RDF would find themselves unable to trackback posts made on new-MT weblogs.
That’s not talking about the very way trackback clients work: appending arguments to a trackback URL is far from being as efficiently simple as doing an XMLRPC call…
2002 09 23
( These news fresh from the blogite mailing list. )
Many discussions on the correct French word to use for ‘weblog’ these days:
- on Blogroots
- on Niutopia
- Michel Dumais’ opinion
- Karl’s logo
- the tall red head‘s lick on the matter
- Stéphanie‘s Google study
As for me ? ‘Weblog’ is the word I’ll use in French.
2002 09 21
Gah, Blake/Martin just beat Llodra/Santoro. It was the first time I watched a tennis match in a while and it seems like it wasn’t the right one to watch despite the very exciting way that both team played. I’m gonna root for Grosjean and Clément to take France to the Davis Cup’s Finals tomorrow, if the resident family doesn’t insist so much on watching the usual sunday’s Chuck Norris junk series.
Talking about Americans, Dan Ross brings back some of my declining faith in a more moderate American way to deal with these recurring middle-east poop slingers Sharon and Arafat. That I wasted about 2 hours on Little Green Racists’ comments section yesterday makes the visit to Dan’s website even more enjoyable. I’d definitely pass the link to Anil if I wasn’t almost sure that he already found it.
2002 09 20
And so it seems that any publicity is good publicity, for Yale and Columbia, for porn and for chicken.
It’s only some days after I registered Trillian Pro that I notice someone ported Gaim to Windows.
Looks like I’ll have to finish my trillian-2-gaim.php buddy list converter, to support Yahoo and MSN contacts, and then I’ll give it an extensive try.
I’m not regretting my purchase of Trillian Pro, though: I would have donated some day anyway, and I like the simple RSS 0.9x reader.
( By the way, since I gave you my phone number, I might aswell give you my IM details: ICQ is 67438124, MSN is firstname.lastname@example.org, Yahoo is cafelog. Please note that even though I wrote b2, I’m ignoring any IM support request about it: this is why there are support forums. )
2002 09 19
What’s weird is that it actually took an hour.
I mean: <big><font face=”BUY A CLUE” color=”red”> DUH </font></big>
I could poop dozens ‘modules’ like that, especially when I could impose my view on the blogosphere and promote my products.
So let’s see, we have three elements, that are supposed to be common to weblogs.
- blogRoll : OK, many weblogs do have a blogRoll, many of these weblogs’ authors don’t bloody care about OPML for it, though
- mySubscriptions : please cite me at least ten non-Radio weblogs that have an OPML file for their subscriptions, keeping in mind that your average Joe-blogger doesn’t know or care about having subscriptions, let alone in OPML format
- blink : it’s common to Mark Pilgrim and Dave Winer, so if you consider one weblog with 2.5 millions page reads can qualify as many weblogs (there are enough page reads to satisfy some hundreds Joe-bloggers, after all), then yeah it’s common to many weblogs, even though it really is common to two weblogs
So I’m all for adding modules to RSS 2.0, but please, let’s name them accordingly. This module could be better nicknamed RadioChannel than blogChannel, seeing how its elements are so specific to Radio Userland.
But I forgot we have to promote our own tools and formats when we devise a module, teehee.
2002 09 18
No, I have only one phone number, I’m at home all the time at the moment. Yes, just one number. That’s right, I have no life.
I should say, I too have only one phone number at the moment, and the problem is that the line is used to connect this computer to the internet (I shall from now on refer to ‘this computer’ by its name ‘Zengun’, to proove my geekiness).
(And if you, reader, want to leave a message (please !) on my answering machine, here’s the number: +33 4 95340914)
From debugging came improving, and now not only is b2′s pingback fixed, but it can now handle the X-Pingback header, and links done to the blog with ‘www.’ or not (it used to break if you didn’t link the right way).
I’m considering making it support non-absolute pingback server URL but that wouldn’t follow the pingback specs, which tells the pingback server URL should be absolute. It would be great if the specs accepted relative URLs, for I don’t see much reasons not to accept relative URLs.
Looks like I broke b2′s pingback yesterday.
The worst problem is that I can’t figure out where the problem is: when I try to pingback a b2 blog from a b2 blog, I got greeted with “The URL you provided doesn’t seem to belong in this site”, which tells me the error is with the pingback server, but when I ‘manually’ do the pingback request everything runs smooth and the pingback is registered, which tells me the error is in the pingback client.
Damn, I wanted to implement checking of the X-Pingback header today, and instead I’ll have to debug an undecipherable bug.
2002 09 17
(Update: the method that I’m talking about is now part of the b2 API, as b2.getPostURL.)
I was just thinking about how blog clients such as w.bloggar or BlogApp could integrate a pingback client (for people whose blog tool doesn’t support pingback), and soon I realised it wouldn’t be possible using the Blogger API or just any API (not sure about the MT API though).
Why isn’t it possible ? Because at no moment does the blog client know which URL the post that you just typed is going to reside on. All the information it has, is the post’s ID that is returned by the newPost method.
So I figured the blog client could use the post ID to get the URL…
And there it is, an extension of the Pingback API: pingback.getPostURL.
Here’s the method’s spec: Read the rest of this entry »