Wikitech-l December 2002

wikitech-l@lists.wikimedia.org

41 participants
139 discussions

Start a nNew thread

TeX
by Tomasz Wegrzanowski 29 Dec '02

29 Dec '02

texvc is done, testing found no bugs. Please install it on all production Wikipedias.

6 15

Minor fixes to TeX
by Tomasz Wegrzanowski 28 Dec '02

28 Dec '02

Patch against the most recent CVS. Please install texvc on all production Wikipedias.

1 0

Re: [Wikitech-l] TeX
by Derek Moore 27 Dec '02

27 Dec '02

>Toby Bartels is a troll and I'm not going to listen any more to what he >says. But he may have some valid criticisms that could only make texvc better. Although he seems to be a bit fundamentalist about what texvc should allow in his comments on the TeX testing pages; e.g., "...we really don't need TeX support inline, it should probably not be supported inline. ... Wikipedia's math should stick to English letters in boldface and italic whenever possible, for the widest readability, with exceptions only for things (like [pi]) that are universally rendered in other fonts." Aside from that, my own observations are this: 1) Fonts seem to be over anti-aliased. '+', '-', and '=' signs are all blury, when they should be distinct, straight lines. 2) Fonts seem to be over weighted, i.e., everything looks somewhat bold. 3) Inline math doesn't feel very readable. PlanetMath's inline equations feel a bit more readable than texvc's. Of course, these are rather non-technical criticisms of texvc's font rendering. I'd suggest employing whatever font rendering defaults that PlanetMath is using, as their equations feel much more readable and much less ugly. Also, what about <math> ... </math> conflicting with MathML? If people employ MathML in an article, does texvc check between the <math> ... </math> tags to see whether it's MathML or TeX? Instead of overloading standard tags, maybe it should be <tex> ... </tex>, and then just let people know that Wikipedia only supports a subset of TeX for math purposes only. In any case, that would make parsing of pages faster, as the "Is this MathML or TeX?" logic could be thrown out. The price of that logic may be trivial now, but what about when Wikipedia is serving hundreds of thousands of pages a day? Plus it would make all the texvc input forward-compatible if it's decided someday that fuller TeX support is needed and that <tex> ... </tex> tags should be used. Just my thoughts... Okay, I'm done now, Derek _________________________________________________________________ MSN 8 limited-time offer: Join now and get 3 months FREE*. http://join.msn.com/?page=dept/dialup&xAPID=42&PS=47575&PI=7324&DI=7474&SU= http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_newmsn8ish…

2 1

wikipedia down?
by Jens Frank 27 Dec '02

27 Dec '02

Hi, is wikipedia.org down? I don't get any pages for the last half hour. Symptoms: - When telneting to ports 80 or 22, connection is established, but no response is sent by wikipedia. This is interesting in the case of port 22 since SSHD should send its version string before the client sends its version string. - No response even after 5 minutes waiting. => No DNS timeout based problem. Assumption: - Out of Memory? - Fork bombed? Could someone have a look at it? Best regards, jens frank -

1 0

Re: A few questions
by Sheldon Rampton 26 Dec '02

26 Dec '02

Erik Moeller wrote: >I presume the conference where you heard about Wikipedia was the World- >InfoCon. If so, did you by any chance hear about it on day two from Volker >Grassmuck? He's a friend of mine, and a couple of weeks before the con he >called me and asked me about cool projects in the area of collaborative >journalism and peer-to-peer networks. I told him about Wikipedia, and he >promised to bring it up at the conference. If that's where you heard about >it, I can take at least some small credit for you being here now :-) You're right, it was at the World-InfoCon, from Volker Grassmuck. Actually, though, several speakers at the conference talked about Wikipedia. >HTML anchors are not yet implemented, but this is a project I'd be willing >to tackle. We just need a nifty syntax. How about just including them with section headers? For example, take the following line of wiki syntax: === Geography and Climate === Right now this produces the following HTML: <h3> Geography and Climate </h3> It shouldn't take a lot of work to make the same wiki syntax produce the following HTML: <a name="Geography_and_Climate"><h3> Geography and Climate </h3> This implementation would only make it possible to insert HTML anchors at the beginning of section headers, but that should cover most of the places where people might want to have them. Then to create a hyperlink to the "Geography and Climate" section of the "Russia" article, you'd simply use the following syntax: [[Russia#Geography and Climate]] >Do I understand you correctly when I >assume that when you want a link like [[person:George W. Bush]], you want >to later be able to link to this page by just using [[George W. Bush]], >but use the "person" prefix during page creation to choose a proper >template? In that case, the question would arise what would happen to this >qualifier after the page has been created. Yes, that's true. The specifics of my proposed implementation are half-baked. The broader concept, though, is the idea of being able to object-type articles as a way of imposing some kind of structure. Maybe this is just fundamentally antithetical to the free-form nature of Wikipedia, but I see that someone else is thinking along rather similar lines with the proposal for a "Slotipedia." -- -------------------------------- | Sheldon Rampton | Editor, PR Watch (www.prwatch.org) | Author of books including: | Friends In Deed: The Story of US-Nicaragua Sister Cities | Toxic Sludge Is Good For You | Mad Cow USA | Trust Us, We're Experts --------------------------------

2 1

Re: [Wikitech-l] Re: A few questions
by Derek Moore 26 Dec '02

26 Dec '02

>>>One other way to get the functionality would be to implement HTML anchors >>>in wiki syntax. That way you could link to [[George W. Bush#people]] to >>>get to the specific article section. You could also implement page >>>templates to get a default structure for articles. > >>Are HTML anchors currently implemented in wiki syntax? > >HTML anchors are not yet implemented, but this is a project I'd be willing >to tackle. We just need a nifty syntax. Actually, there really shouldn't need to be any special syntax for anchors. 98% of the time, you'd want an anchor at your headings and nothing more (well, maybe at your media, too). That should be as simple as modifying the wiki syntax parser. For headings, if the heading was "External links" the anchor's name could be something like "#external_links" or "#External_links" or "#External links"... What do you think, should anchor names be lower()'d? Should their spaces be translated to '_' and their punctuation ignored? I tend to think "yes" to both questions. Media embedded in articles could also probably benefit from anchors. Again, no need for additional syntax--it ought to be automagic. Anchor names for media could just be the media's filename. <plug>By the way, for those interested in new and improved wiki syntaxes, check out Wikitax: <http://meta.wikipedia.org/wiki/Wikitax> <http://meta.wikipedia.org/wiki/Talk:Wikitax> Erik, Tomasz, and Nemo have contributed some excellent ideas to the syntax so far, and everybody else interested is encouraged to do the same. I, personally, tend to think the wiki syntax could use plenty of revamping. If you tend to think so, too, stop on by and add your ideas, etc.</plug> Peace out, Derek _________________________________________________________________ MSN 8 with e-mail virus protection service: 3 months FREE*. http://join.msn.com/?page=features/virus&xAPID=42&PS=47575&PI=7324&DI=7474&… http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_eliminatev…

1 0

mod_rewrite configuration
by Derek Moore 26 Dec '02

26 Dec '02

Check this out: http://www.wikipedia.org// I typo'd that earlier. Check it out. I'm sure the above is a fairly common typo, so somebody might wanna fix the mod_rewrite rules to allow for it. Peace out, Derek _________________________________________________________________ The new MSN 8: smart spam protection and 3 months FREE*. http://join.msn.com/?page=features/junkmail&xAPID=42&PS=47575&PI=7324&DI=74… http://www.hotmail.msn.com/cgi-bin/getmsg&HL=1216hotmailtaglines_smartspamp…

4 8

Re: A few questions
by Sheldon Rampton 25 Dec '02

25 Dec '02

Eric Moeller wrote: >Whoa, > >now that's a surprise -- a guy whose work I like to reference on Wikipedia >and elsewhere shows up on, of all places, the Wikipedia tech mailing list. >Welcome, Sheldon, and thanks for your excellent work with John Stauber on >exposing the corporate PR machinery. Thanks! (I'm blushing.) >I've often wanted to have more online material of yours that I could point >people to, and perhaps Disinfopedia can become that. I'm assuming that you've seen the PR Watch web site. We have a fair amount of material there, including archives of PR Watch dating back almost to its founding, plus our "Spin of the Day" feature (which typically includes 10-15 new items each week). I'm envisioning the "Disinfopedia" as an open source extension of our "Impropaganda Review" section, which you can find at the following URL: http://www.prwatch.org/improp/ Once I have the wiki software tweaked and functioning the way I want, I'll seed the Disinfopedia with our existing articles from the Impropaganda Review plus some other similar information, and offer it under the GNU General Public License. At some future point, we may choose to do the same thing with other parts of our web site. > Of course, I would >also appreciate it if you could contribute info, when you find the time, >to the "real" Wikipedia. Have you thought about the license you want to >use for your project yet? It's hard to operate a wiki under traditional >copyright, and more beneficial to society not to. I've already made a few minor contributions to the "real" Wikipedia, and I plan to continue doing so. You've touched here on one of the questions that I've been wondering about: Should we set up a separate "Disinfopedia," or just use the "real" Wikipedia as a repository for the knowledge base I'm trying to develop? On balance, I think setting up a separate "Disinfopedia" is the way to go, for the following reasons: (1) A "Disinfopedia" is likely to be more controversial and contested than Wikipedia, and the differences are significant enough to raise compatibility questions. (For example, Wikipedia's ethos regarding a "neutral point of view" may be harder to maintain in a "Disinfopedia.") (2) There isn't a good way to clearly mark off a "Disinfopedia" as a subset of Wikipedia, and simply posting our material there would dilute and diffuse its specific focus. (3) Having a separate Disinfopedia makes it easier for us to offer guidance on article format and content. For example, I want to encourage contributors to the Disinfopedia to follow the model we used in the Impropaganda Review, where each organizational profile consists of the following sections: * A general description * Personnel * History * Funding * Case studies * Contact information * Related information resources >As for your namespace links: It's possible - I would try to reduce the >number of topical namespaces to 3 or 4 to avoid ambigiousness ("where do I >need to go to get information X? case studies? issues? hmm .."). Also, try >to think about good names. People:George W. Bush? Is that a page about >George Bush or a page with links to people related to George Bush? (IMHO >only the latter makes sense, as the page about George Bush should not have >its own namespace, but this is not intuitively clear.) Excellent points. I'm envisioning the extra namespaces as places for lists of related links, so People:George W. Bush would be a page with links to people related to Bush. Also, you're right that we don't need a separate namespace for "issues." Perhaps the thing to do would be to simply have ONE additional namespace, titled "relationships" (plus, of course, "relationships_talk"). At the bottom of an article on "George W. Bush," there would be a link to an article on Bush's "Relationships to People, Organizations, Funders, etc." Assuming the article hasn't yet been created, the URL would read: http://www.disinfopedia.org/wiki.phtml?title=Relationships:George_W._Bush&a… Clicking on that page would take people to a textedit form, but instead of the start text saying "Put your text for the new page here," it would contain a template, e.g.: >>Lists of people, events, funding, case studies and information >>resources related to [[George W. Bush]]. >> >>== People == >> >>* [[Name1]] >>* [[Name2]] >> >>== History == >> >>* [[Event1]], date >>* [[Event2]], date >> >>== Financial == >> >>* [[Funder1]] >>* [[Funder2]] >> >>== Case Studies == >> >>* [[Article1]] >>* [[Article2]] >> >>== Documentation == >>* [[Author]], [[Title]], [url], name of publication, pub date, page Yet another possibility would be to forego extra namespaces altogether, and simply post prominent guidelines explaining the kind of structure we're envisioning. I would hope that people would TRY to follow our recommended structure, but it would be okay if they sometimes deviate from it. >One other way to get the functionality would be to implement HTML anchors >in wiki syntax. That way you could link to [[George W. Bush#people]] to >get to the specific article section. You could also implement page >templates to get a default structure for articles. Are HTML anchors currently implemented in wiki syntax? Also, how do you envision implementing page templates? Are you thinking of something along the lines of the example I gave above? And what about the idea of creating "pseudo-namespaces"? What I'm envisioning here is a way of object-typing prospective articles so that page templates could be customized according to object type. For example, [[person:George W. Bush]] might create a template with sections such as date of birth, career highlights, etc., whereas [[organization:Cato Institute]] might create a template with sections such as personnel, funders, publications. Has there been much discussion about ways of integrating Wiki's free-form approach with XML-style object-typing and data-structuring? >PS: "Toxic Sludge is Good for You" should really be available as a GNU FDL >e-book. We need to get this book into schools :-) Personally, I'd be interested in seeing this happen, but "Toxic Sludge" was published by a commercial press, and I don't know how they'd feel about releasing it into GNU space. On a related tack, I've been toying with the idea of writing my NEXT book as an open source project from start to finish. We would begin by posting a chapter outline and letting anyone edit, contribute and annotate while we write. Do you think this would work? Would it be possible to publish the book commercially and get it into bookstores if the writing process took place within a GNU framework? Would the GNU license be necessary to attract collaborators? -- -------------------------------- | Sheldon Rampton | Editor, PR Watch (www.prwatch.org) | Author of books including: | Friends In Deed: The Story of US-Nicaragua Sister Cities | Toxic Sludge Is Good For You | Mad Cow USA | Trust Us, We're Experts --------------------------------

2 1

Re: A few questions
by Sheldon Rampton 25 Dec '02

25 Dec '02

Derek Moore wrote: >Tehehe, me too, actually! I was pretty impressed for wikitech-l when >Sheldon appeared. You's a pseudo-celebrity amongst radicals, Sheldon. Great! Does that mean I get groupies? ;) Wikipedia is a natural place for me to show up. For one thing, I've always been a bit of a computer hacker. (I founded the computer club at my high school and got suspended from school once for hacking into a university computer so I could learn COBOL.) Also, my last book with John Stauber was titled "Trust Us, We're Experts: How Industry Manipulates Science and Gambles With Your Future." We did a lot of writing and thinking about the scientific method and the influence of funding on research outcomes, and also about the nature and limitations of academic peer review. I recently adapted part of "Trust Us" for a medical journal, and when they asked me to offer solutions for some of the problems that we posed, the best answers I could come up with were: (1) better standards of disclosure regarding funding and other possible conflicts of interest related to published research; and (2) maintaining and expanding an "information commons" in which research results are treated as public property and kept freely available to everyone. One of the big problems for the integrity of research these days is that corporate-sponsored research tends to produce proprietary knowledge, which inhibits full publication of data, methodology and other information. Companies have good reasons from their point of view for not wanting to release discoveries that they have paid for and from which they hope to profit. However, there have been cases where companies have deliberately suppressed data showing that their products are unsafe or ineffective. Even where this sort of misconduct isn't occurring, restrictions on proprietary information present a barrier to full vetting and sharing of research results. Believe it or not, I actually thought I was coming up with a new idea when I first used the term "information commons." I was familiar with the concept of open source software, but I hadn't really given the concept much attention other than to think it might be nice if the same approach could be adapted to things other than computer software. Then last month I attended a conference in Amsterdam where I heard about Wikipedia for the first time and thought, "Damn! These people are already doing it!" -- -------------------------------- | Sheldon Rampton | Editor, PR Watch (www.prwatch.org) | Author of books including: | Friends In Deed: The Story of US-Nicaragua Sister Cities | Toxic Sludge Is Good For You | Mad Cow USA | Trust Us, We're Experts --------------------------------

1 0

Re: Proposal: make summary mandatory
by Giskart 25 Dec '02

25 Dec '02

[to:wikien-l, to:wikitech-l] Axel Boldt wrote: > I would like to propose that we make filling out the summary field > mandatory on ever edit. Reasons: > > * it makes the job of vandals and vandal bots harder, since they have > to come up with a meaningful summary each time. > * it makes the history page of an article vastly more useful; now I > often find myself employing the bisection method to find out when a > particular bit of information was inserted. Filling in the summary field is in general a good thing to do, but sometimes I find it stupid because the work it put in writing the notice for the summary field is more work than the change of the article. A pulldown menu whit some default comments seems a good way. More options to follow close the actions of not registerd users would be very nice. So that you can put a namelabel to a IP-adress, for those IP-users who you can trust be have the bad idea to not login. You can also use it to label proxyservers, so you see directly in RC that a ip-adress is a proxyserver. Now you have to look every ip up, but it are mostly the same who come back and give trouble. > I posted this to wikien-l, since this feature would best be implemented > as an option that other language wikis could switch on if desired. I think all Wikipedia's would like more options to better view what is happening. Giskart

2 1

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Wikitech-l December 2002