Older blog entries for cdent (starting at number 420)

SomeInfo201206

Date Users Spaces Created Modified Tiddlertime Searchtime
20120601 5272 8130 163 273 590.1 168.86
20120602 5276 8135 90 168 316.5 1685.82
20120603 5281 8141 154 263 312.3 1819.95
20120604 5287 8154 253 519 318.1 1702.66
20120605 5289 8161 160 252 316.4 1880.38
20120606 5297 8171 241 314 310.7 1933.73
20120607 5308 8183 355 609 311.1 8711.58
20120608 5316 8193 1232 1439 308.5 1837.46
20120609 5321 8202 273 343 316.5 839.28
20120610 5323 8204 129 235 328.1 1704.39
20120611 5333 8217 273 391 311.4 1830.38
20120612 5337 8224 259 410 317.8 1886.09
20120613 5344 8233 1233 1349 313 1920.06
20120614 5351 8236 212 371 311 1947.6
20120615 5357 8243 178 241 320.8 1759.63
20120616 5359 8246 90 403 311.2 1773.95
20120617 5367 8255 189 269 531.1 1738.64
20120618 5371 8259 187 319 321 217.89
20120619 5380 8270 445 513 308 1750.61
20120620 5384 8275 309 494 318.2 1806.74
20120621 5396 8291 547 706 321.7 2025.37
20120622 5406 8305 1015 1114 321 13003.6
20120623 5409 8308 67 179 300.8 1782.84
20120624 5412 8313 113 206 317.4 1934.67
20120625 5416 8320 1100 1977 309.3 1763.91
20120626 5429 8337 324 454 525.8 2565.18
20120627 5434 8343 209 363 459.3 1786.76
20120628 5441 8354 326 442 327.6 1889.3
20120629 5451 8365 217 315 318.9 47302
20120630 5453 8370 102 181 311.7 1764.33

Syndicated 2012-07-04 14:29:34 from cdent

httpbin(1): HTTP Client Testing Service

httpbin(1): HTTP Request & Response Service
httpbin(1): HTTP Client Testing Service

Useful?

Syndicated 2012-07-04 14:28:19 from cdent

3 Jul 2012 (updated 9 Jul 2012 at 10:16 UTC) »

StoryFodder 201207

There are plans to do an osmosoft iteration of TiddlySpace development for two weeks in July. What follows are some ideas from which we might be able to generate stories. Similar writing from other people (osmosoft and not) will help shape the iteration.

Doing this instead of writing actual stories for the sake of brainstorming. Many of these are in the form "find the bugs in subsystem X and fix them" which aren't necessarily what stories are for but may be what tiddlyspace needs.

  • Audit and refactor /_space and /_account. Both have several (unrecorded) bugs and inscrutable code.
  • Ditto takenote.
  • Redo default HTML rep CSS to be aligned with the styles used in backstage, apps, space, account, etc.
  • Create a simple search interface to go in universal backstage that uses /hsearch. The purpose is to encourage discovery in fuzzy ways (which /search is less good at). The output would need to be more verbose than a standard tiddlers collection, but have similar style.
  • Do a documentation sprint and only a documentation sprint (yes, for two weeks, with snacks).
  • Simplify, clean and check for correctness the downloaded TiddlyWiki. Make it more like a plain tiddlywiki that could sync to a TiddlyWeb server that happens to be TiddlySpace.
  • Redo the frontpage now that pre-existing constraints are gone.
  • Make eastmad pick some stories.
  • Make an interactive demo that demonstrates the reply/following/atbox ideas.
  • Create a styleguide for TiddlySpace, align existing CSS etc accordingly.
  • Make tsapp.
  • Move bookmark stuff into core.
  • Collapse system-* stuff into one space whereby including or discluding it turns "tiddlyspaceness" on or off in a TiddlyWiki.
  • Allow signup via github or twitter or facebook. That is: figure out oAuth.

It's difficult because to a certain extent many of these things aren't worth doing if the existing system is not:

  • polished
  • self-describing

Historically we've been shit at both those tasks and the story/iteration pattern hasn't serviced us well for accomplishing them. It may be better to just say that for the duration of the iteration people will actually work on (and use!) TiddlySpace rather than their several other responsibilities, with an eye to polish and documentation. It's a drag and a chore and often boring, but it has to be done.

From Others


Summaries or responses of comments by others:

StoryFodder 201207 from pmario has two main desires:

  • The "buttons" on the front page must work. Important part of documentation.
  • Documentation sprint.

Manifesto Review 201206 from fnd:

  • More complete support for Markdown.
  • Analyze public tiddlers to learn of "real" usage patterns.

StoryFodder 201207 from bengillies ends with:

  • Improve the documentation.
  • A consistent style guide ... some work done to put it in place
  • A coherent story for ... the social stuff

StoryFodder 201207 from colmbritton seems to agree with my assertion that we need to just use it:

People should use it, we should discuss our experiences using it, we should fix bugs and we should polish stuff and finish things off, maybe documenting stuff as we go.

This remains my conclusion: We should commit to focusing on using and improving TiddlySpace for the next two weeks, with the unbiased input of Patrick and (the new) Ben and just do the slog to get through things. What's been lacking in the past has been committed and focused time.

The common theme seems to be documentation, attention to detail and polish, and understanding and exposing the capabilities of the system.

Syndicated 2012-07-03 16:27:30 (Updated 2012-07-09 10:16:16) from cdent

3 Jul 2012 (updated 3 Jul 2012 at 16:10 UTC) »

20120703

Pushed out TiddlySpace 1.1.6 which engages some of the ts.js changes discussed yesterday.

bengillies found a search bug and I fixed it.

Syndicated 2012-07-03 13:14:19 (Updated 2012-07-03 15:57:56) from cdent

TiddlySpace 1.1.6

  • use status.js in some apps
  • adjust ts.js to use status.js
  • Cloning an included tiddler should respect the chkPrivateMode
  • csrf time fixes

Syndicated 2012-07-03 13:13:44 from cdent

2 Jul 2012 (updated 2 Jul 2012 at 21:12 UTC) »

20120702

Released a new tiddywebplugins.status that provides a /status.js akin to the one suggested in this thread. The resulting data is found in a tiddlyweb.status object.

This ought to make certain types of widgets for the HTML rep a bit easier. Thanks to bengillies for bringing the idea out in conversation.

I'll be updating some of my widgets today to make sure it is behaving as expected.

I've updated edit so that editedit.js will use tiddlyweb.status if present and edit itself uses /status.js directly.



A big drive for making this change was limiting calls to /spaces/<spacename>/members that are only used to determine the current membership status of the current user in the current space. The main culprit of this behavior is the UniversalBackstage which does it because ts.js does it.

That code is desperately entwined with that concept. Besides being used in the backstage it is also used by /_space,/_account and the registration and login processes on the front page. That this is so is exactly the kind of messy complexity that the status.js changes are trying to improve.

So I'll go digging in that code and see what I can do.

Um, wow. Complexity. ts.js is trying to do too much for too many different things. When the backstage loads, it appears that it attempts to gather up the information that it might need were it actually the _space or _account applications. I'm exploring to see if I can get things a bit more clear, but my initial reaction is that we should audit:

  • use of ts.js
  • use of all the various chrjs libraries

I'm sure there are plenty of refactoring opportunities, given what we've learned since the code was first created.

I've made a pull request on ts.js with some initial thoughts.

Syndicated 2012-07-02 14:06:52 (Updated 2012-07-02 20:34:12) from cdent

PyCon UK 2012 - Eventbrite

Back by popular demand, PyCon UK is the place for UK Pythonistas to meet, preach and learn.
PyCon UK 2012 - Eventbrite

I suppose I'll probably go to this.

Syndicated 2012-07-02 11:22:13 from cdent

z2012-06-28- Tiddly Space Progress - WebSeitz/wiki

Maybe it's a good time to build an RSS-based Open Twitter crossed with a WikiLog that supports Sister Sites!
z2012-06-28- Tiddly Space Progress - WebSeitz/wiki

Syndicated 2012-07-02 11:17:49 from cdent

411 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!