I've just committed a bunch of new stuff to CVS:
1) "Edit this section". If you enable the user preference "Show links for editing individual sections", you get little "edit" links under each article section. These can be used to fetch just the text of that section, and edit it. No more scrolling through 30 K articles to edit a typo. No more wading through long discussion threads to add a short comment (provided they are organized using headlines).
2) Automatic table of contents. If the option "Show table of contents for articles with more than 3 headings" is enabled, a small TOC is added on top of the page with navigation links to the individual sections. From this it follows logically that we now have
3) Anchors for each article section (named after the section title, e.g. "External links" becomes "External_links"). So you can link to these from elsewhere.
Regards,
Erik
Erik Moeller schrieb:
I've just committed a bunch of new stuff to CVS:
Great, thanks!
- Anchors for each article section (named after the section title, e.g.
"External links" becomes "External_links"). So you can link to these from elsewhere.
Just to be sure, "from elsewhere" means "from elsewhere inside the same article", right?
Kurt
Kurt-
- Anchors for each article section (named after the section title, e.g.
"External links" becomes "External_links"). So you can link to these from elsewhere.
Just to be sure, "from elsewhere" means "from elsewhere inside the same article", right?
From wherever you want. Say you have a headline "External links". Now you
can say
[[#External links]]
for an internal reference, or
[[Wikipedia#External links]]
from elsewhere. It should work with all existing headlines, although you have to be sure your cash is clear.
Regards,
Erik
Erik Moeller schrieb:
Kurt-
- Anchors for each article section (named after the section title, e.g.
"External links" becomes "External_links"). So you can link to these from elsewhere.
Just to be sure, "from elsewhere" means "from elsewhere inside the same article", right?
From wherever you want.
But we have discussed this over and over. Please don't do it.
Kurt
On Mon, 30 Jun 2003 14:07:17 +0100, Kurt Jansson jansson@gmx.net gave utterance to the following:
Erik Moeller schrieb:
Kurt-
- Anchors for each article section (named after the section title,
e.g. "External links" becomes "External_links"). So you can link to these from elsewhere.
Just to be sure, "from elsewhere" means "from elsewhere inside the same article", right?
From wherever you want.
But we have discussed this over and over. Please don't do it.
Without reaching consensus. It's a feature I very much want. Remember - the number of "views" of an article will probably outnumber the number of edits by 100 to 1. Isn't a little bit of inconvenience for editors outweighed by a lot of convenience for readers who aren't editing?
Richard Grevers schrieb:
But we have discussed this over and over. Please don't do it.
Without reaching consensus. It's a feature I very much want. Remember - the number of "views" of an article will probably outnumber the number of edits by 100 to 1. Isn't a little bit of inconvenience for editors outweighed by a lot of convenience for readers who aren't editing?
But we're not talking about "a little bit of inconvenience", but about * a which-articles-link-to-this-anchor feature (so you know which articles to change after restructuring) (* this might be easier by some kind of Redirect for anchors that aren't used anymore) * a most-wanted-anchor feature (to keep track of articles that lead to deleted anchors)
To me this sounds frightening, and I'll leave restructuring of articles with anchors to others. And I guess I'm not the only one.
Kurt
On Mon, 30 Jun 2003, Kurt Jansson wrote: [snip]
To me this sounds frightening, and I'll leave restructuring of articles with anchors to others. And I guess I'm not the only one.
While I'm not very enthused about the idea of using volatile header text as anchor names, I'm not nearly so worried about having anchors as you:
* If an anchor name changes, the link still takes you to the wrong part of the right page, just as it does now.
* If an anchor-linked section is removed from an article and/or made into a separate page, any anchor-links to it will fall on the old page, exactly as they would now without the anchor link.
* Identifying pages which link to some particular anchor is no more difficult than identifying pages which link to the page intending to point the reader at some subsection but without being able to do it. Programmatically it could be made much much easier, if someone cares to make that additional feature.
Of course, maybe we *shouldn't* have pages so long that people feel the need to anchor into them. But if we do, *AND WE DO RIGHT NOW*, is it a crime to simplify the lives of people trying to read those pages at the cost that, occasionally, a link might break and be no worse than it is at present?
-- brion vibber (brion @ pobox.com)
Brion Vibber schrieb:
While I'm not very enthused about the idea of using volatile header text as anchor names, I'm not nearly so worried about having anchors as you:
Sorry if I seem a bit heated - I'm not :-)
Of course, maybe we *shouldn't* have pages so long that people feel the need to anchor into them. But if we do, *AND WE DO RIGHT NOW*, is it a crime to simplify the lives of people trying to read those pages at the cost that, occasionally, a link might break and be no worse than it is at present?
Okay, I'm still sure that we'll end up in a complete mess (or loose much time to keep everything tidy, or maybe both), but I won't argue against it anymore. I'll just bookmark a link to this mail and wait some months ... :-)
Kurt
Erik Moeller wrote:
- Anchors for each article section (named after the section title, e.g.
"External links" becomes "External_links"). So you can link to these from elsewhere.
If by "from elsewhere", you mean �from other articles�, then am I the only one left that considers this a Bad Idea?
But if by "from elsewhere", you mean �from elsewhere in the same article�, then great!
-- Toby
Toby-
If by "from elsewhere", you mean «from other articles», then am I the only one left that considers this a Bad Idea?
Well, as I recently learned, it was always possible to link to anchors within other articles using the [[foo#bar]] syntax. Only the anchors were not yet added to the articles (but could be added manually using HTML). Now the anchors are auto-generated, because it was necessary for the table of contents generation to do that. I'm not inclined to modify the code to take a previously existing feature (the [[foo#bar]] support) out, which means that it is now possible to link to sections on any page from any page. Call it a side effect that was difficult to avoid.
Regards,
Erik
Erik Moeller wrote:
Toby Bartels wrote:
If by "from elsewhere", you mean ?from other articles?, then am I the only one left that considers this a Bad Idea?
Well, as I recently learned, it was always possible to link to anchors within other articles using the [[foo#bar]] syntax. Only the anchors were not yet added to the articles (but could be added manually using HTML). Now the anchors are auto-generated, because it was necessary for the table of contents generation to do that. I'm not inclined to modify the code to take a previously existing feature (the [[foo#bar]] support) out, which means that it is now possible to link to sections on any page from any page. Call it a side effect that was difficult to avoid.
I wouldn't call that a previously existing /feature/, since it was useless without an HTML trick that was never intended. That it becomes useful is a side effect of what you've done, yes; but then, its very existence is just a side effect of how we parse links. However, since you didn't /create/ this, let me rephrase:
I suggest disabling intrawiki links of the form [[X#Y]] (but not [[#Y]]), for the reasons that may be found in past discussions.
I am still generally favorable, of course!
-- Toby
Toby-
I suggest disabling intrawiki links of the form [[X#Y]] (but not [[#Y]]), for the reasons that may be found in past discussions.
Yes, we can do that. I'd like to have a general consensus on the matter, though. Heh, just kidding. But we should have a vote first. Several people have requested this functionality.
Regards,
Erik
--- Erik Moeller erik_moeller@gmx.de wrote:
Toby-
I suggest disabling intrawiki links of the form [[X#Y]] (but not [[#Y]]), for the reasons that may be found in past discussions.
Yes, we can do that. I'd like to have a general consensus on the matter, though. Heh, just kidding. But we should have a vote first. Several people have requested this functionality.
Did I miss the general consensus or vote that allowed your patch in?
I am against inter-article anchor links because they make the rewriting and refactoring of long articles much harder, yet that will soon be the most important if not most common activity on Wikipedia.
Axel
Regards,
Erik _______________________________________________ Wikitech-l mailing list Wikitech-l@wikipedia.org http://mail.wikipedia.org/mailman/listinfo/wikitech-l
__________________________________ Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com
On Mon, Jun 30, 2003 at 10:57:26AM -0700, Axel Boldt wrote:
--- Erik Moeller erik_moeller@gmx.de wrote:
Toby-
I suggest disabling intrawiki links of the form [[X#Y]] (but not [[#Y]]), for the reasons that may be found in past discussions.
Yes, we can do that. I'd like to have a general consensus on the matter, though. Heh, just kidding. But we should have a vote first. Several people have requested this functionality.
Did I miss the general consensus or vote that allowed your patch in?
I am against inter-article anchor links because they make the rewriting and refactoring of long articles much harder, yet that will soon be the most important if not most common activity on Wikipedia.
I think that's not the right attitude. Techinal people should implement whatever may be useful, even when there are downsides, and consensus should be reached about policy of usage of these new features. This consensus may be different for different Wikipedias, and almost for sure it will be different for Wiktionary (where there are standard anchor targets "X Language" and multipart articles). That way every wiki has what it wants. Mixing policy issues with technical issues enforces views of majority on everyone else, even if different projects have drastically different needs.
Tomasz Wegrzanowski wrote:
On Mon, Jun 30, 2003 at 10:57:26AM -0700, Axel Boldt wrote:
I think that's not the right attitude. Techinal people should implement whatever may be useful, even when there are downsides, and consensus should be reached about policy of usage of these new features. This consensus may be different for different Wikipedias, and almost for sure it will be different for Wiktionary (where there are standard anchor targets "X Language" and multipart articles). That way every wiki has what it wants. Mixing policy issues with technical issues enforces views of majority on everyone else, even if different projects have drastically different needs.
Unfortunalty, everytime someone say something like: have a different default skin, someother say's: not without consens over all wikipedias. Can you imagine thier reactions on your proposal?
There seems to be no way of 'every wiki get's what it wants' -- theres only one solution possible for all wikipedias - unfortunatly.
Tomasz Wegrzanowski wrote:
Axel Boldt wrote:
Did I miss the general consensus or vote that allowed your patch in?
I think that's not the right attitude. Techinal people should implement whatever may be useful, even when there are downsides, and consensus should be reached about policy of usage of these new features.
You're correct, but I think that some discussion may still be useful. For example, is it possible to disable [[X#Y]] links on Wikipedia but enable them on Wikitionary, or would this be horrid to code? A justification for allowing these links in Wiktionary is that Wiktionary has highly standardised header names; they're always "English", "French", "German", "Polish", etc. But headers in Wikipedia are unstable.
Not that this is the only possible solution, but a few more days of discussion could help, as long as people agree that we're working towards compromise. In the end, I'd suggest erring on the side of too much functionality and relying on policy to hold this back -- but that's imperfect too, since it's hard for policy to stop people from doing what they can. I'll advertise this on <wikipedia-L>.
I also have another technical question for Erik: How does [[Special:Whatlinkshere]] interact with this?
-- Toby
Erik Moeller wrote:
Toby Bartels:
I also have another technical question for Erik: How does [[Special:Whatlinkshere]] interact with this?
It simply shows the links to the page, without the anchor. But the anchors don't break Special:Whatlinkshere.
I'm afraid that I don't understand the response. '_` I'll ask a more precise question:
If [[Z]] links to [[X#Y]] but not to just plain [[X]], then will [[Special:Whatlinkshere&target=X]] list [[Z]]?
-- Toby
Axel-
Did I miss the general consensus or vote that allowed your patch in?
I have already explained this, but I will gladly explain it again: My patch did not change the linking behavior. All it did was add automatic anchors to headlines, not because I wanted to allow interpage-anchor- linking, but to implement a table of contents that is shown on top of long articles (more than three headings). This was suggested here before by precisely the people who *opposed* interpage-anchor-linking. Unfortunately or not, by having anchors in the page for the TOC, it is now possible to link to these anchors from other pages as well. The code for doing so was already there, apparently since the first version of Wikipedia Phase III. If it is to be taken out, that should be discussed first.
Regards,
Erik
Erik Moeller schrieb:
The code for doing so was already there, apparently since the first version of Wikipedia Phase III. If it is to be taken out, that should be discussed first.
But it has been, over and over and also right after LDC implemented it. He said he wouldn't activate it, but it seem it hast activated itself. I can't find the old mails through google. Why isn't our archive searchable? Dammit, let's put an advertisement on /. and k5, "Wp needs more programers!" or something like that. :-)
Kurt
Kurt-
maybe that's a silly question, but why don't we define a policy on interpage-anchor-linking? There may be a few instances in which it is useful (FAQ? link to "External links" section, which is always titled the same anyway?) without any drawbacks.
Regards,
Erik
On 6/30/03 5:05 PM, "Erik Moeller" erik_moeller@gmx.de wrote:
Kurt-
maybe that's a silly question, but why don't we define a policy on interpage-anchor-linking? There may be a few instances in which it is useful (FAQ? link to "External links" section, which is always titled the same anyway?) without any drawbacks.
This whole discussion REALLY should be taking place on the wikipedia-l list.
Unilateral, unilateral.
Cunc-
Kurt-
maybe that's a silly question, but why don't we define a policy on interpage-anchor-linking? There may be a few instances in which it is useful (FAQ? link to "External links" section, which is always titled the same anyway?) without any drawbacks.
This whole discussion REALLY should be taking place on the wikipedia-l list.
It will, once people can actually try out the feature. Until then, moving it there makes little sense. It's not even on test.wikipedia.org yet.
But hey, nothing like stirring up things a little, eh?
Regards,
Erik
Erik Moeller schrieb:
Kurt-
maybe that's a silly question, but why don't we define a policy on interpage-anchor-linking? There may be a few instances in which it is useful (FAQ? link to "External links" section, which is always titled the same anyway?) without any drawbacks.
Or we could handle them like subpages and only allow them in non-article namespaces. Either way I'll have to explain to newbies why they can't/shouldn't be used in articles. But okay, if everybody can agree on this.
Kurt
Kurt Jansson wrote:
But it has been, over and over and also right after LDC implemented it. He said he wouldn't activate it, but it seem it hast activated itself. I can't find the old mails through google. Why isn't our archive searchable? Dammit, let's put an advertisement on /. and k5, "Wp needs more programers!" or something like that. :-)
More programers to produce more code waiting in the CVS for better days? Try to think about your suggestion again.
Thomas Corell schrieb:
More programers to produce more code waiting in the CVS for better days? Try to think about your suggestion again.
Okay, I hereby solicit to give more people access to the server. People I personally trust not to do anything against the will of the community (and that hopefully will keep the German Wikipedia up to date ;)) are e.g. Erik, Thomas and Magnus.
And, could someone please set "Los" (Go) button default when you hit enter in the search box on the German Wikipedia, disable SQL queries or do anything else to speed the German Wikipedia up. It was often near to unuseable in the last afternoons. We got a very nice article in a journalism magazine and we shouldn't give such a bad impression to the visiting journalists. I hope some of them will soon write about the project in other magazines ...
Every new feature should be put on hold until proven that it doesn't put more burden on the server. We should roll out the red carpet for everybody who promises to help with the database problems. Welcome Timwi! ;-)
Kurt
Hi,
Okay, I hereby solicit to give more people access to the server. People I personally trust not to do anything against the will of the community (and that hopefully will keep the German Wikipedia up to date ;)) are e.g. Erik, Thomas and Magnus.
Who does have access to the German Wikipedia server currently?
We should roll out the red carpet for everybody who promises to help with the database problems. Welcome Timwi! ;-)
Hahaha! :-) ... I haven't promised anything yet, though. :-)
Greetings, Timwi
Kurt Jansson wrote:
Thomas Corell schrieb:
More programers to produce more code waiting in the CVS for better days? Try to think about your suggestion again.
Okay, I hereby solicit to give more people access to the server. People I personally trust not to do anything against the will of the community (and that hopefully will keep the German Wikipedia up to date ;)) are e.g. Erik, Thomas and Magnus.
I would like server access to work on the skins & stylesheets. I can keep language files updated too (I've learnt from the HUGE mistake I made last time!)
If some tells me where I can find the code that handles image markup, I may look into that too.
I'd like to approve Erik, Magnus, and Tarquin for server access, if that is something they'd like to do, and if that's something that we think would be useful.
There are probably others who I've overlooked, but if lack of people with direct server access is a bottleneck, we should do something about that.
In general, even developers don't really need direct server access. As long as we have enough people to update the live site from CVS at appropriate times, and enough people to give the machines a kick in the pants when needed...
Direct server access is a matter of great trust, not just in terms of character, either. Lots and lots of people have excellent character and I would trust them on those grounds with server access. But there is a 'technical' component to it, i.e. knowing how to get around in Unix without breaking anything.
And I guess there's a personality component, since the server setup isn't like a wiki where lots of people should be on there willy-nilly reconfiguring things. :-)
But it seems like this is sort of a bottleneck, and the solution is pretty simple.
--Jimbo
Jimmy Wales wrote:
I'd like to approve Erik, Magnus, and Tarquin for server access, if that is something they'd like to do, and if that's something that we think would be useful.
As developer I can only see the duration changes need from the CVS to the live wiki. Only the persons doing this job know if manpower is missing. I hope it helps :)
There are probably others who I've overlooked, but if lack of people with direct server access is a bottleneck, we should do something about that. In general, even developers don't really need direct server access. As long as we have enough people to update the live site from CVS at appropriate times, and enough people to give the machines a kick in the pants when needed...
I agree with that.
Direct server access is a matter of great trust, not just in terms of character, either. Lots and lots of people have excellent character and I would trust them on those grounds with server access. But there is a 'technical' component to it, i.e. knowing how to get around in Unix without breaking anything.
Well, this will be seen usually very fast. And it's like making 'sysops' in the wiki's, if it fails remove them fast. In the hope only persons agree to do the job, have appropriate experiences.
And I guess there's a personality component, since the server setup isn't like a wiki where lots of people should be on there willy-nilly reconfiguring things. :-)
But it seems like this is sort of a bottleneck, and the solution is pretty simple.
*keepingfingers crossed*
In general, even developers don't really need direct server access. As long as we have enough people to update the live site from CVS at appropriate times, and enough people to give the machines a kick in the pants when needed...
Well, but developers would need write access to CVS then.
Timwi
wikitech-l@lists.wikimedia.org