| |
|
Wednesday, 19 June 2002
|
|
The Register: Microsoft Restores Java to XP. Here's the anatomy of Microsoft's decision-making.
- In an effort to cut off Java's client-side air-supply, Microsoft removes their JVM from XP.
- Sun milks every drop of publicity out of this. Anyone who finds an applet they can't use is able to download Sun's JRE1.4 Java plugin, which supports all the latest APIs, and has nifty (but as yet unrealized) features like Java Web-Start
- With this, plus Netscape/Mozilla's out of the box support for the modern Java plugin, Applet technology starts looking like it might finally be able to escape the doldrums caused by the fact the major browsers never upgraded to Java2.
- Microsoft realises they made a mistake, and re-packages their ancient not-quite-JDK1.1 plugin, making it look like they've capitulated to Sun's whining, but actually restoring the old nightmare, and creating new inertia against modern Java applications.
12:40:19 PM
|
|
|
Thursday, 13 June 2002
|
|
Step one: a replacement Radio aggregator that remembers when you've un-checked a story, and doesn't check it again. This hack will only be useful for people who have this preference turned on..
License: This code is provided for free. It is derived from the aggregator code within Radio Userland, I thus retain no rights to it. It is UNTESTED CODE, and is provided WITHOUT WARRANTY OF ANY KIND. If it corrupts your data, crashes your computer, erases your hard drive, sleeps with your wife and performs satanic rites on your dog, don't come crying to me.
Instructions: Put the root file somewhere Radio can see it. Edit www/system/pages/news.txt and replace the macro call with one to aggregatorCategoriesSuite.viewNewsItems (). The aggregator should behave pretty much like it used to, it'll just remember when you uncheck boxes.
Note: What I did could be accomplished with about five lines of changes to radio.root, rather than an entirely new tool. The reason it's a tool is that this is the first small step in a pretty big overhaul of the aggregator, I'm just releasing it now because it's doing something useful already.
The aggregator might run a little slower than before. I did a lot of refactoring of the Userland code and moved a bunch of logic around in order to separate the "select which stories to display" code, the "handle the user request" code and the "draw the page" code.
11:15:50 AM
|
|
|
Wednesday, 12 June 2002
|
|
Found in radio.html.viewNewsItems:
- (local flSkip = true)
- if adritem^.url == xmlUrl
- if flskip
flskip isn't used anywhere else. I am now officially afraid of reading Userland code. :)
2:55:53 PM
|
|
Dave Winer responds, but his response makes my head hurt. We seem to use one definition of "journalist" when talking about professionals, and another definition when talking about webloggers, but this proves webloggers are journalists. Back when I did Philosophy 100, this was called equivocation.
If declaring your interests up-front is a valid way out for webloggers-as-journalists, then it must be for journalists-as-journalists. Otherwise, we have to declare the two to be fundamentally different ideals.
Journalism for self-interest is called publicity. Perhaps we're not journalists, we're self-publicists.
12:23:15 PM
|
|
Radio News Aggregator features I want, in order of importance:
- Multiple categories for RSS feeds, each with its own aggregator page
- The ability for news items to remember that they'd previously been marked "don't delete", so I don't have to keep un-checking their delete boxes as I go through the rest of the list
- The ability to navigate past the first page of items
The first is the most wanted. Categories in the aggregator would make my life so much more pleasant, and make my news-reading much, much more efficient.
I am subscribed to about 20 news sources of varying volumes, and the news still gets on top of me. With the three features above, I could probably subscribe to an order of magnitude more.
11:04:14 AM
|
|
Matthew Thomas indirectly poses a question.
Assume most "tech bloggers" are employed. This puts them in a position to have detailed knowledge of the field in which they work, including the goings-on within their employer, and the exact position of the competition in their market. This also puts them in a position to be in really big trouble if they disclose that information. If part of the definition of a journalist lies in the editorial freedom to criticise one's employer, can bloggers be journalists?
For example, you work for vendor X. Your employer has published benchmarks, but you know the truth, your product does X and Y better than anyone else (and you're proud of that), but people probably shouldn't use it for W and Z. This is information that really should be given to the Real World. And devotees of the Cluetrain Manifesto (or at least that subset of it that I agree with wholeheartedly) would tell you that making the knowledge public would only increase peoples' confidence in your product, because they finally had some obviously honest advice.
But if you say so in your weblog, marketing is going to nail your ass to the wall.
Bloggers' employers are generally not media companies, and so are not at all pre-disposed to grant employees any leeway for public criticism. A journalist may get away with writing that their boss lost a few million dollars in a stupid deal, a programmer wouldn't.
On top of that, I've lost count of the number of NDA's I'm subject to at the moment. Every project I go on, I'm under some kind of non-disclosure agreement, so if I were to start blogging about things that were happening at work, I might be open to serious liability.
There's a reason I don't name my employer in my blog. I'll disclose that we're a small IBM business partner doing Java development in Sydney, and that we're home to a large number of Davids. You may work out who we are from that, but if you do, you probably know me already.
10:51:32 AM
|
|
|
Tuesday, 11 June 2002
|
|
Ick. Whenever I try to go to my prefs page, Radio crashes.
2:50:42 PM
|
|
|
Friday, 7 June 2002
|
|
Brunching Shuttlecocks: The Weblog FAQK
Gosh Jesus no! Weblogs cover a wide range of topics, such as other weblogs, what the mainstream media are saying about weblogging, new weblogs, advances in weblog publishing, books about weblogging, the future of weblogging, and that one naked guy painted up like Spider-Man.
3:51:28 PM
|
|
|
Thursday, 6 June 2002
|
|
A quick note for anyone writing documentation, or filling in default values in the configuration file for an application.
The domains example.com, example.net and example.org are reserved for use in documentation. Thus, if you use bob@example.com in your default configuration file, you can be guaranteed that the address will never exist, and thus there will never be a "Bob" who is annoyed at your stupid choice of example domain names.
Addendum: The authority for this comes from RFC 2606, which also reserves the following top-level domains for the various, obvious uses: .test, .example, .invalid and .localhost.
4:42:09 PM
|
|
|
Wednesday, 5 June 2002
|
|
A quick huzzah to my TiBook. Her name is epiphany (or 'epi' for short). I just upgraded to OS X 10.1.5, and must reboot. According to 'uptime', my last reboot was 29 days ago, also occasioned by an OS upgrade. Not bad for a laptop, eh?
Of course, my desktop W2K box has a similar uptime. And my Linux box has been up 70 days now. This is why I tend to stick anyone I see who says "I tried [Operating System], but it crashed all the time!" in the this-person-is-full-of-shit basket. (for modern values of [Operating System]).
3:53:11 PM
|
|
Davezilla: These are the Daves of our Lives. "Everyone knows a Dave or three. Daves are always dependable, competent, rather silly and the jack-of-all trades in most offices. Dave is always the guy who can fix the copier, jumpstart your engine or make that noisy dog calm down." [Scripting News]
When I started working for my current employer, the company had five Davids (including two of the three company directors). Which wouldn't be too surprising, except that there were only around ten people in the company.
2:44:15 PM
|
|
A link that might be useful to web-designers, from Mark Pilgrim:
Adrian Roselli: A Simple Character Entity Chart. Both entity names and entity numbers are listed, along with the representation of the character (so you can test for compatibility by loading this page in your browser).
2:04:32 PM
|
|
|
© Copyright 2002 Charles Miller.
Last update: 27/6/02; 8:44:09 AM.
|
|
| June 2002 |
| Sun |
Mon |
Tue |
Wed |
Thu |
Fri |
Sat |
| |
|
|
|
|
|
1 |
| 2 |
3 |
4 |
5 |
6 |
7 |
8 |
| 9 |
10 |
11 |
12 |
13 |
14 |
15 |
| 16 |
17 |
18 |
19 |
20 |
21 |
22 |
| 23 |
24 |
25 |
26 |
27 |
28 |
29 |
| 30 |
|
|
|
|
|
|
| May Jul |
|
|