Archived sections

Put new stuff below here

User talk:Short Circuit/Image Upload problem

Contributions question

What does "top" mean in the "my contributions" list ? I think I could't figure it out so far! Ulrie

I think it means there haven't been any edits more recent than yours. --Michael Mol 21:14, 6 September 2010 (UTC)

I think it might mean that you edited the entire page instead of just a section (as in you clicked the edit link at the top of the page). It doesn't have much significance anyway I don't think. --Mwn3d 21:50, 6 September 2010 (UTC)

Nevermind. Mike is right. --Mwn3d 21:51, 6 September 2010 (UTC)

Oops!

I uploaded a new version of [this image]. I accidently upload a 3mb version. How do I delete the image so it isn't hoarding storage space? Chris Ferri 19:47, 12 August 2010 (UTC)

Let me worry about bandwidth and server storage space. The server made thumbnails, so as long as those are what get embedded in various places, things should be fine.

That said, if you want to replace the file, try the "upload new version" link that's on that page somewhere. --Michael Mol 19:49, 12 August 2010 (UTC)

Hm. That image could probably use a run through pngcrush, though. GIF might work as a better format, too; 8-bit palettized, with LZMA compression, works nicely for some scenarios. Play around with it. :) --Michael Mol 19:51, 12 August 2010 (UTC)

K, thank you for the help. Chris Ferri 20:19, 12 August 2010 (UTC)

LabVIEW

So, for LabVIEW, the optimal solution is to actually upload the vi containing the solution code rather than upload an image of the code. Would that be OK, or would that consume too much storage space? LabVIEW vi's can be anywhere from 7KB to several megs. --Chris Ferri

Hm. That's actually a very, very interesting problem. First, MediaWiki (what this site runs on) limits allowed upload file types. I can modify that, but that wouldn't be a good policy habit to get into. Additionally, a 7KB block of source code would be a rather large segment to inline into a page--and MW can OOM on multi-megabyte pages. (That's assuming the .vi files were human-readable.) So while uploading the .vi files is the optimal solution from a code execution standpoint, it's not practically workable. That leaves screenshots, assuming there's enough information in the screen shot for the user to build the solution.

Now you've got me seriously pondering something like mercurial as a back-end for holding the source code--that would efficiently handle the scenario. It would also require a massive redesign of how the site software works.

Can screen shots work? (I think you're replying to something, but I don't know off-hand where the other part of this conversation took place.) --Michael Mol 14:56, 14 July 2010 (UTC) 11:17, 14 July 2010 (UTC)

If I could chime in on the subject, I think it would be a good idea to have both the image and the vi file itself. The image would give a general idea of what the program looks like to someone who's not familiar with or does not own LabVIEW. But that alone could potentially make it very tedious to try. Some LabVIEW programs can grow to be quite large and could be time-consuming to re-write. A downloadable version could be a huge time saver for those who own a version of it. So I would recommend having an image and a download link, if possible. --Tyrok1 16:14, 14 July 2010 (UTC)

Really, I agree. The problem, though, is a technical one; I don't have an effective means of allowing the upload and storing of .vi files. --Michael Mol 16:18, 14 July 2010 (UTC)

Ok well below are the front panel and block diagram for heap sort implemented in LabVIEW. This vi in my opinion represents the average labVIEW vi. My problem with this is the only way to see the other cases and nested functions etc would be to upload an image of them as well. This could lead to potentially many dozens of images needed to show all the cases in a case structure etc. And, anybody wanting to use this code won't have the ability to run the code without all of these images from which to recreate these functions. I have no idea how your file hierarchy is setup, but one solution to the storage problem would be to create a "labVIEW" folder in which you can place all files with a .vi file extension that are uploaded.--Chris Ferri

The least tractible problem stems from implementation limitations: MediaWiki file uploads are limited by PHP settings on the server, which are in term adjusted with an eye towards resource usage (most notably memory, but disk usage is something I'd be worried about as well.). The best practicable solution I could provide would be a sideband upload mechanism such as a source code repository that used MediaWiki credentials for authentication. Doable, and has interesting and useful side effects wrt the rest of the site, but that'd take planning. I actually had a face-to-face conversation with Tyrok1 today, where he explained to me some of the limitations of the screenshot approach, so I understand your problem. I just don't have the time to implement it right now. --Michael Mol 22:39, 14 July 2010 (UTC)

No probs. It's not a huge issue for me to be unable to upload the LabVIEW solutions. I'm more interested in staying in practice and solving the tasks on this site give me a good way to do that. But, when or if this site is ever capable of handling LabVIEW solutions to programming tasks, I'd happily take the time to upload all of them. Until then, I see no point in uploading screen shots of the code without the source to go along with it (for people who use labVIEW). As you can see, even trivial algorithms are difficult to decipher without labVIEW's profile and debug tools.--Chris Ferri

Has anyone asked the company that creates LabVIEW if they have a way to textually dump its programs for use on sites such as this? I can see that a non-graphical, but still 'meaningful' textual form might be useful to others.(They could output a version in graphviz dot format for example). --Paddy3118 05:06, 15 July 2010 (UTC)

Not AFAIK, but I did learn that the .vi file contains positional data for blocks and connections, that the blocks are generally very simple, that most meaningful programs wind up constituted of a huge number of blocks, and that all that data for all those very simple blocks adds up. I don't know how much it supports by way of linked-in libraries and code-reuse. If there were a textual format, I'd be surprised if some third-party tools for library processing and such didn't crop up. --Michael Mol 14:36, 15 July 2010 (UTC)

There is no way to textually represent the G source code. It would be pointless, because the source for a labVIEW vi not only contains the positioning info for the visual programming elements, but also hidden labVIEW processes running in other threads. It would be harder to decipher than a picture of the block diagram. So, I do have the option of uploading the vi's to the LabVIEW Developer Zone, and then link to them from there. But, I am hesitant to do so because I don't know what license user created programs are licensed under. --Chris Ferri

Multilingual

Are you going to ever introduce a multilingual in Rosetta Code?

A what? --Michael Mol 18:46, 8 April 2010 (UTC)

Like into Wikipedia. Sorry, i.e. adding multi-language. --DCamer 18:52, 8 April 2010 (UTC)

I'm still not clear on what you're trying to say. Is English your best language? If not, try giving a full description in a language you're comfortable with, and I'll throw it through, e.g. Babelfish and possibly another native speaker. Might be easier that way. --Michael Mol 18:55, 8 April 2010 (UTC)

Do you plan to add to the site other languages except English? (comfortable to me: Планируете ли вы добавлять на сайт другие языки кроме Английского?). --DCamer 19:01, 8 April 2010 (UTC)

Ah, I see what you mean! I don't know how practical that would be on MediaWiki. The source code itself should be common between all human languages, even if the individual descriptions and the rest of the pages would not be. I don't think it will happen while the site runs on MediaWiki, but it's definitely something I would plan to allow for if/when a suitable alternative can be implemented. --Michael Mol 19:06, 8 April 2010 (UTC)

You could do it via subdomains like Wikipedia does -- like en.rosettacode.org vs ru.rosettacode.org. -- Eriksiers 19:10, 8 April 2010 (UTC)

That's not the problem. The problem is that the core content, the lang-by-task matrix of examples, is reasonably correct for all languages, but the supporting description and explanation isn't. The way the examples are currently presented on the site, we would see the example code for any lang-task point diverge, without an effective way to resolve updates and differences between the various languages. (Granted, even with code fixedly common between all human languages, you still run into quirks of variable names and comments. Adding locale support to source code presentation would be a daunting (though interesting) task in itself!) --Michael Mol 19:40, 8 April 2010 (UTC)

I see. Yes, you're right, Mediawiki is probably unsuitable as-is. Oh well. -- Eriksiers 20:22, 8 April 2010 (UTC)

So there is extension Interwiki and manual :). I tried on the local machine - no problem. --DCamer 21:26, 8 April 2010 (UTC)

We have InterWiki, but the problem is that we'd want to transclude the code itself across all languages, because the code itself should be common to all human languages. There are two problems with transclusion. First, the code examples would need to be separated from page content (which I already believe needs to happen, personally), with the English and Russian versions of the site transcluding from them. Second, with the transclusion, it's not trivial to make a change to a code example; you would have to navigate to the sourced page and make the change there. (If that second part weren't a problem, I would have already addressed the first part by breaking out the code examples into the "Example:" namespace I've already created on the site.) --Michael Mol 21:34, 8 April 2010 (UTC)

But why the code to separate from the main text? So that the changes in the code would be reflected simultaneously in all languages? --DCamer 10:33, 9 April 2010 (UTC)

Yes, exactly. Otherwise, it becas a conflict resolution problem much like that of source control, but without the aid of automated toos. --Michael Mol 11:23, 9 April 2010 (UTC)

Then it is possible to create the code in the various article and to refer to it through the template. {{Example:Code1}}. --DCamer 11:55, 9 April 2010 (UTC)

Translated from earlier, Google Language Tools: Есть две проблемы с transclusion. Во-первых, примеры кода, должны быть отделены от содержимого страницы (о котором я уже считаю, должно произойти, лично), с английской и русской версии сайта transcluding от них. Во-вторых, transclusion, это не тривиально внести изменения в пример кода, вы должны перейти на страницу источников и внести изменения там. (Если это вторая часть не проблема, я бы уже обратились к первой части, разбив из примеров кода в "Пример:" имен я уже создал на сайте.) - Майкл Мол 21:34, 8 апреля 2010 (UTC) --Michael Mol 12:00, 9 April 2010 (UTC)

Alternatively - you could create it through a template, here an example, bottom of the page, or on website MediaWiki. --DCamer 19:36, 8 April 2010 (UTC)

Trying again. (I passed it backward and forward through Google until it came back saying what I was trying to say.): Есть две проблемы с transclusion. Во-первых, образцы должны быть отделены от содержимого страницы (о котором я уже считаю, должно произойти), с английской и русской версии сайта transcluding от них. Во-вторых, это не тривиально внести изменения в теле шаблона от страницы, использующие этот шаблон, надо перейти на страницу шаблона и сделать один изменим там. (Если второй половине не было проблем, я бы уже implemanted первую часть, разбив примеры кода из в "Пример:" пространство имен, что я уже создал на сайте.) --Michael Mol 18:36, 9 April 2010 (UTC)

As you wish, however, often themselves examples contain of comments and text strings, that should also be translated, so it turns out that problems with sample no. If the project has become multilingual and international, it was be cool! :)
P.S. I apologize for bump post. --DCamer 20:52, 15 April 2011 (UTC)

No problem. I wish we could do it, but it'd be a _lot_ of maintenance and overhead to be done properly. I know there's a massive following from your area, but I don't have the resources to do the thing properly. --Michael Mol 20:56, 15 April 2011 (UTC)

Excuse me, but about what resources are you talking? Money? To attract the Staff? Or to technical side? With respect to the second, they are not needed, just a subdomain, and multiple directives in LocalSettings.php. With respect to Staff: the people there is, are able to control the language section. --DCamer 11:41, 16 April 2011 (UTC)

The resources to get the wiki switched over to a semantically structured and constructed layout. See Rosetta Code:Village Pump/Semantic MediaWiki#Discussion, particularly the part where it talks about using Hough Transform as a test environment. That's the single largest thing that makes this difficult to do. If that were taken care of, most of the rest of the maintenance headaches would vanish, and the rest manageable, as long as there were some dedicated folks to help maintain the RU side of internationalization. (I also expect there to be friction and communications difficulties across internationalizations with respect to task clarification, but that's a problem that can be addressed once it happens) --Michael Mol 13:26, 16 April 2011 (UTC)

In addition, in the interwiki there is «transwiki transclusion», allows the use pages from the other wiki as templates. --DCamer 12:06, 16 April 2011 (UTC)

Rcode buttons

I dig it. Nice touch. --Mwn3d 01:51, 11 April 2010 (UTC)

Slowly gearing towards prettifying things and improving site structure and layout. I think I need to use a better font, though. And I want the server to generate and cache the alternate sizes. And I want MW to handle size specifications in points, darnit! Each button is a 48k image, currently, but at 96x96px the Language button is a mere 8.4k, before and after pngcrush. --Michael Mol 01:58, 11 April 2010 (UTC)

AssaCom is spam user

I squelched the content of User:AssaCom because it was spam (it was apparently talking about Russian airlines according to Google Translate). Since that was content put there by the user, it was almost certainly a user created solely for spamming and you'll want to squelch the account ASAP. –Donal Fellows 15:16, 16 April 2010 (UTC)

Pipe Dream

Is it possible to have a "Recent changes" for a category, or a selection of categories? Ideally, I'd like to be able to see the changes in Category:J or Category:Programming tasks, since I last logged in. Is that even feasible?

--DanBron 23:25, 16 April 2010 (UTC)

Not with the current software set. If you can find the appropriate MediaWiki extension, I'd probably add it. (I myself have wanted a feed specific to the Talk: namespaces for a long time.) --Michael Mol 23:40, 16 April 2010 (UTC)

Actually, using the MediaWiki API, you could filter recent changes however you liked, the only real limitation being that the API doesn't let you look back in time any further than Special:RecentChanges does. —Underscore (Talk) 00:20, 17 April 2010 (UTC)

[Solved] Rewording required?

In several places I can see this phrase: These tasks are not considered "unimplemented". Wouldn't it be better to say: These tasks are not considered "unimplementable". Or is it just a misunderstanding on my behalf? (English is not my first language). Wolf 11.May 2010 12:00h (GMT+2)

It says what it means, but the problem is that I wasn't talking about whether or not a task is implementable, but whether or not it counts towards a set of "unimplemented" tasks—a question of pride for some people. --Michael Mol 13:17, 11 May 2010 (UTC)

I understand now, thanks. I have provided some code for the task "Sort stability" using the AutoHotkey programming language. The task shows up OK on the page for implemented tasks for AutoHotkey. How do I remove "Sort stability" from the list of "Not Considered" tasks, please? Or is that up to you to do?

Somewhere on that task page will be a reference to Template:Omit, with an argument of AutoHotKey. Find that bit and remove it. --Michael Mol 16:34, 11 May 2010 (UTC)

All done. Thank you again. Wolf 11.May 2010 18:51h (GMT+2)

Time constriction ?

Sorry, to bother you again. I have written some code for the task "Arbitrary-precision_integers_(included)" in the programming language AutoHotkey. Now my problem here is the following: the code runs on my PC for 26 hours and then spits out the correct result. I have used an implementation of BCD multiplication by AHK forum user Laszlo (permission has been granted to publish his code) together with my own implementation of "long powers", that held tight when tested with Euler Challenges. I am as sure as I can possibly be, that this code is correct (forum discussion starts with the second post from the top here: AHK forum, if you want to have a look). Now I think that pride thing you mentioned yesterday has got a grip on me too, so here is my question: Is there a time constriction at all for submissions to RosettaCode? I assure you that this is not some untested, half-baked code that gets thrown at RosettaCode, but it can be improved. Wolf 12.May 2010 14:32h (GMT+2)

No time constriction. --Michael Mol 13:42, 12 May 2010 (UTC)

But isn't a condition of that task that it is only solvable by languages that come with their own in-built implementation of arbitrary precision integers? If you are having to code it then it should not be included. --Paddy3118 14:09, 12 May 2010 (UTC)

You're correct. And this is a discussion better suited for that task's talk page. --Michael Mol 14:22, 12 May 2010 (UTC)

Captcha's back?

Is the presence of the captcha a conseguence of something I am doing or whatever or it is back also for registered users? --ShinTakezou 17:05, 7 June 2010 (UTC)

It should be back for everyone, but only when someone makes an edit that adds an offsite link. Spambot registration is getting through the initial CAPTCHA, most likely as part of a people solver farm, but it's not as worth it for such attacks to solve CAPTCHAs for every spam edit. Sucks, but there it is. There might be an alternate solution, using random numbers of CAPTCHAs for registration, but that'd require modifying the SimpleCAPTCHA and/or ReCAPTCHA extensions. I can create a "skip captcha" privilege group, if people like, but it would require someone to go through and confirm users after one or two valid edits. --Michael Mol 17:49, 7 June 2010 (UTC)

Personally, I would eventually like to become part of that privileged group, if you have the time to implement it. Specifically, I can no longer use the presence of captcha to remind me that I am not logged in at submit time, and I would like to get that back. But I have no problem waiting for 100 valid edits and 30 different days where I edit content (or whatever other filtering mechanism seems reasonable), if you want to automate it. --Rdm 18:25, 7 June 2010 (UTC)

I think all currently-registered users could be put in that group, regardless; any spam accounts are already banned. Implementing the privilege group is technically easy, too. The problem comes from the need for a human judgement for new accounts. --Michael Mol 18:31, 7 June 2010 (UTC)

I should note that currently I am required to fill out the recaptcha form even when I am signed in, and have added no links and the text of the section I am editing contains no links. --Rdm 15:02, 8 June 2010 (UTC)

Ew. I'm surprised it's doing that. I'd rather deal with a couple more spammers. Went in and checked settings, changed some things. Let me know if it's still a problem. --Michael Mol 15:08, 8 June 2010 (UTC)

I have made several edits, and (including this one) none have required I fill out the captcha form. Thank you! --Rdm 15:55, 8 June 2010 (UTC)

URLs should still require it, FYI. --Michael Mol 15:57, 8 June 2010 (UTC)

Note, however, that now I do not need captcha to submit even when I am logged out (though I have not tried posting any URLs that way). --Rdm 12:04, 9 June 2010 (UTC)

Avoiding spam URLs are the key issue, but I did like having all anon edits bring up CAPTCHAs. Don't have time to fix this week. Later. --Michael Mol 12:28, 9 June 2010 (UTC)

Same for me, no ext link afair, but captchaed anyway. Last edit went smooth captchalessly, thanks —ShinTakezou 18:41, 8 June 2010 (UTC)

Spammer?

Content of http://rosettacode.org/wiki/Rosetta_Code_talk:Village_Pump looks spammy, like a test run for blog spam. Particularly telling is the fact that the author hasn't contributed anything else. You might want to just revert things back there and have a think about whether some other form of block should be applied… –Donal Fellows 12:16, 10 June 2010 (UTC)

I'll take a wait-and-see approach. Individual users can be mass-rolled-back if necessary. I can see about re-enabling CAPTCHAs for all non-logged-in, but I really don't have the time right now to deal with anything short of an emergency situation. --Michael Mol 13:40, 10 June 2010 (UTC)

Back to

I've tried to fix the page Pascal's triangle from irriverent edits by user Robertom, but noted too late that s/he deleted some implementation while adding his/her one! I think it is better to "revert" the page to the version before the Robertom's edits, and eventually add APL impl later. I believed I could do it by myself, but it seems I have no the right, or simply too (something) now to understand how I can do it ... I don't know who can, so writing here (maybe village pump would've been a better place?). Thanks --ShinTakezou 11:57, 11 June 2010 (UTC)

I believe I have restored the missing edits (I restored AutoHotKey, BASIC and C). Basically, I just went back to older versions where they were there and copied and pasted them into the current verison. --Rdm 13:46, 11 June 2010 (UTC)

Problems with uploading new versions of existing images

Looks like there's a problem with uploading new versions of existing images. See File talk:Matlab-randomDisc-output.png and File talk:Eriksiers avatar.jpg. (Might this be related to the recent upstream problems? Seems pretty unlikely, but you never know...) -- Erik Siers 17:42, 22 September 2010 (UTC)

Might be related to the switch to the addition of Squid. Don't know. Don't have time to look at it right now. I'll check the logs this weekend, if I have time. --Michael Mol 18:29, 22 September 2010 (UTC)

As long as you're aware of it, I'm happy...ish. :-) -- Erik Siers 18:43, 22 September 2010 (UTC)

Looked at the PHP error log file. I'm not seeing any file I/O errors, or anything in the vicinity of your avatar replacement log entry's timestamp, but I am seeing a number of "Maximum execution time of 30 seconds exceeded" errors relating to SMW activity over the past few days. I might be able to resolve some of that. --Michael Mol 18:48, 22 September 2010 (UTC)

Tweaked a couple variables. Bumped max_execution_time to 45s, reduced the job run rate to 0.1. Not a permanent fix for the timeout issues. Need to fix squid, and deal with some serious performance issues in SMW. --Michael Mol 18:52, 22 September 2010 (UTC)

I've noticed some occasional-but-significant delays in loading pages today, but didn't really give it much thought... but it could be the cache, I guess. Would make sense, I suppose. -- Erik Siers 19:05, 22 September 2010 (UTC)

The cache won't catch nearly as much as would be ideal, because of complexities surrounding X-Vary-Options. The significant delays you see are going to be cache misses combined with other load on the server. (It looks like the SMW export functionality is getting hit pretty hard.) --Michael Mol 19:12, 22 September 2010 (UTC)

Ok. I'll hold off uploading again until you have time to sit and fiddle with everything (and I know how precious time can be). It's not very important to me but I can see it being a problem that might potentially affect everything... or... something. -- Erik Siers 19:44, 22 September 2010 (UTC)

TBH, I don't have a timeframe in mind for a proper fix for the caching issue. It'll involve using a patched version of Squid, and I really can't anticipate when I'll have time to deal with that. Apart from that, updating MediaWiki (I think I'm a version or two behind by now) may help.) A workaround might be to delete the page and recreate it. You'll have to poke someone with the relevant privs. --Michael Mol 19:51, 22 September 2010 (UTC)

Again

Me again. Looks like you never got around to a permanent fix on the cache problem; I uploaded a new version of my user avatar and it's still showing the old version on my user page (checked on a few machines on different networks). I know essentially nothing about caches so I can't offer any help beyond "hey, not working".

(My only thoughts on the matter may or may not be feasible; it would involve getting Mediawiki to somehow tell Squid to clear the cached version of an updated image, but that would likely require some hacking at the mw sources. Maybe the Squid sources, too, dunno.) -- Erik Siers 04:39, 4 February 2011 (UTC)

After a hard refresh (ctrl+R in Firefox) I saw the new version (the Feb 1 version). Have you tried that? --Mwn3d 04:47, 4 February 2011 (UTC)

I had done that without effect, right before I posted (and numerous times after uploading), but now that I look again, it's showing the correct one. WTF? -- Erik Siers 04:54, 4 February 2011 (UTC)

MediaWiki and Squid don't get along terribly well, I'm afraid. MediaWiki does have code for telling Squid to purge caches, but it requires a specially-patched version of Squid. The patch had been submitted to Squid devs ages ago, but they declined to include it in trunk. The key, though, is that the cache object is tied to a literal match of your browser's supported-encoding string. --Michael Mol 13:58, 4 February 2011 (UTC)

That's... kinda weird. Is it one of the patches in here? I might take a look at it (if I can understand it). -- Erik Siers 14:36, 4 February 2011 (UTC)

this one. --Michael Mol 14:58, 4 February 2011 (UTC)

Contest with RC questions

It seems an APL programming contest used RC questions: here.

Yeah, I first noticed referral traffic from Dyalog's contest a few months ago. Pretty cool. :) --Michael Mol 09:03, 27 September 2010 (UTC)

Template work

So I just started some work on the templates and I noticed TONS of templates like this one: Template:Unimp body 4D. They were made in the before time (in the long, long ago) by ImplSearchBot. Are we not using them anymore? Can I delete them? --Mwn3d 14:57, 11 October 2010 (UTC)

Check to see what pages transclude them, and see if anyone links to those. That info should be available in the toolbox section in the sidebar. If they're not used, wipe 'em. :) --Michael Mol 15:06, 11 October 2010 (UTC)

Stats

Thanks for the welcome.

I was wondering if there were any analytics available on the different languages and tasks. How many tasks are completed for each language? For a given task, how many languages implement it (and how many ignore it)?

Of course any individual question can be answered with a bit of counting, but to see larger trends isn't so easy. For example: of the dozen or so languages where I could conceivably write a solution, which are most underrepresented? Similarly, which tasks seem important yet aren't widely solved?

I toyed with the idea of writing a screen-scraping bot, or even writing a task to write such a bot, but I thought I'd ask here first.

CRGreathouse 04:38, 3 November 2010 (UTC)

Actually, it looks like that is mostly covered by Rosetta Code/Count examples and Rosetta Code/Find unimplemented tasks. Maybe add extra credit for a visualization? CRGreathouse 04:59, 3 November 2010 (UTC)

Each language is a category so you can get (a good approximation to) the info directly on its page. You can also get the other information from the language's report page (e.g. for Tcl, the page is Reports:Tasks not implemented in Tcl). –Donal Fellows 10:57, 3 November 2010 (UTC)

Attack page?

Template:Unimpl Page/Not implemented and Category:Programming Tasks were both recently, uh, 'vandalized' by User:Umobytuz who linked them to http://osobageqys.co.cc and http://evicijum.co.cc which seem to be attack/exploit pages. You may wish to check on the IP address of this contributor and/or check for other similar recent additions.

I removed both links.

CRGreathouse 06:38, 24 November 2010 (UTC)

Thanks for the clear-up and notice. I took a look then blocked the user immediately. If I was too harsh, then I'd rather SC unblock than leave the guy easy access. --Paddy3118 07:09, 24 November 2010 (UTC)

No, blocking isn't too harsh. That means it's probably past time to apply active protections on templates and transcluded pages, too. --Michael Mol 14:43, 24 November 2010 (UTC)

User:ShariPantoja‎ appears to be another spammer (using their talk page). –Donal Fellows 11:43, 25 November 2010 (UTC)

Did some WP financial research

I just spent a little time digging through the Wikimedia Foundation's site, trying to find some info on how to get money from them. What results I got I put here --> Rosetta Code:Village Pump/Income#Dear Wikimedia Foundation <-- in case you missed it on the "recent activity" thing.

Not much helpful info, but I tried. Possibly you've already seen everything I found anyway. Shrug. -- Erik Siers 07:23, 3 March 2011 (UTC)

...and now something else you might ponder: Rosetta Code:Village Pump/Income#Paid Memberships. -- Erik Siers 16:00, 3 March 2011 (UTC)

RC used in a talk by Larry Wall (of Perl)

... And reviewed+blogged by me --> User_talk:TimToady#On "That Goes Without Saying (or Does It)".

Squid configuration

It looks like Squid's maximum request size is causing problems for AutoGeSHi. —Underscore (Talk)

Spammer

User:Debonairlazines6974 is a spammer (check their contributions). Alas. –Donal Fellows 12:08, 12 May 2011 (UTC)

Got it covered now. Thanks Donal. --Mwn3d 13:06, 12 May 2011 (UTC)

I've made a right mess

Mike. I made a mistake with a rename, and I tried to fix it, but the fixes keep digging me deeper and deeper into a hole. Can you back out my changes from 19:29 to 19:48 today (11th July). Cheers. Markhobley 19:53, 11 July 2011 (UTC)

Not automagically, it looks like. The 'undo' and 'rollback' features look a little more complicated for page moves that page-local edits. At work right now, and will be busy this evening. I'd suggest popping into (and hanging out in) the IRC channel and enlist some assistance and coordination there. --Michael Mol 19:56, 11 July 2011 (UTC)

Book Mention

In "A Byte of Python". --Paddy3118 12:21, 17 July 2011 (UTC)

Cool. :) --Michael Mol 12:25, 17 July 2011 (UTC)

More financial suggestions

...this time not from me, but posting here in case you missed it:

-- Erik Siers 12:46, 4 October 2011 (UTC)

Saw it then. Just been generally busy, and coping with a hardware failure. Slowly resolving these things...--Michael Mol 23:25, 14 December 2011 (UTC)

Spam users

I happened to look at the user creation log, and holy crap look at all the apparently-created-to-spam usernames -- things like "4Hjyo29LKb", "Gzs6na7KyZ", "4LwiEd2p", etc. Looks to be on the order of 5-10 spam accounts to 1 legit account.

I was thinking, there has to be a way to catch those, before they start posting their links. The only method that comes to mind is blocking their entire subnet, at least temporarily, say 30 days. That way, they can't just come back and create a new username and continue spamming, or just spam without logging in.

Might be harsh, especially if they're spamming from a normal ISP account, but maybe a short-term solution. Would likely reduce the amount of admin work, methinks. -- Erik Siers 20:20, 4 October 2011 (UTC)

They do get their source IPs blocked, and the IPs are disallowed from creating new users. I don't have a mechanism that allows me to block entire subnets, but that will become necessary when the site moves over to dual-stack IPv4/IPv6. --Michael Mol 20:32, 4 October 2011 (UTC)

And before anyone suggests it, I'm not interested in applying pattern-matching or bayesian tests on usernames. I've been caught by surprise before when an account that looked like a spam username started making beneficial edits. --Michael Mol 20:32, 4 October 2011 (UTC)

What about one of those tests that simply sends an alert to an admin? Make it a one-time thing, something like: "SpamWatchBot has noted an edit from suspicious user name apparent-spammer-name on page So-and-so. This is the only alert that will be generated for this user."

Then the bot (or whatever) could add apparent-spammer-name to an internal ignore list, and let the admins worry about it.

Eh... Looking at it now, maybe not worth the time to write. Shrug. Or maybe there's already a MW bot available to do something similar and could be modified appropriately; I haven't looked. Shrug again. -- Erik Siers 07:15, 7 October 2011 (UTC)

Right now, I see almost every edit on the wiki, as I've got the Recent Changes feed pulled in via Google Reader. Anyone who does that can see when accounts are created or blocked, and when pages are edited or moved. Frequently, someone else gets in and repairs things before I have a chance to respond. I wind up blocking the offending account, possibly deleting any 'created' pages if the person who caught the edits first didn't have delete privileges. --Michael Mol 12:52, 7 October 2011 (UTC)

Uploading problem

I've been trying to upload a file and the interface won't let me. It's a png file which I've cunningly disguised with a .png extension, but when I attempt to upload it I just get

Permitted file types: png, gif, jpeg, svg.

Any ideas?

I had written an elegant solution to a task but I daren't post it without its output since I get harassed whenever I have the temerity to post a solution without giving its results.

CRGreathouse 07:15, 23 November 2011 (UTC)

This is a known problem. See Image Upload problem? ~~and Problems with uploading new versions of existing images~~ on this page. -- Erik Siers 07:31, 23 November 2011 (UTC)

Ah. No solution known, then? I didn't see one at either of those links. CRGreathouse 08:22, 23 November 2011 (UTC)

The one I crossed off is probably unrelated. No response from Short Circuit so I assume no solution yet. He had a combination server move/upgrade planned; I don't know what happened with that but I guess that maybe he was hoping that the problem would disappear when he did that. If he did that. -- Erik Siers 08:27, 23 November 2011 (UTC)

There's some related discussion on my talk page, see if it's of any help. --Ledrug 08:51, 23 November 2011 (UTC)

There is a workaround linked to in the above discussion: [1]. --Mwn3d 11:57, 23 November 2011 (UTC)

I believe it to be related to "AJAX" uploads. That's what I found when I tried walking through the JS code. What I don't get is why people are still seeing that message even after I disabled AJAX uploads. Sorry for not replying directly in the other places it's cropped up; I'm beginning to think we need a real bugtracker for this stuff. --Michael Mol 14:03, 23 November 2011 (UTC)

I managed to make it work by disabling JavaScript on my browser. CRGreathouse 03:12, 28 November 2011 (UTC)

The MediaWiki software has been updated, serverside. Is this still going on? --Michael Mol 23:24, 14 December 2011 (UTC)

<math> not working?

Equations in the tasks are not being rendered for some reason. Could you check this? Thanks. --Paddy3118 07:00, 14 December 2011 (UTC)

Also broken on all other pages; for example, “<math>1</math>” here renders as

1

(which used to render as “1” because the code used to spot that it was simple enough to convert to HTML; its definitely not getting as far as trying to send the info into TeX and failing). I guess it was the weekend's server changes, and could be as simple as a missing handler for the <math> tag. –Donal Fellows 09:18, 14 December 2011 (UTC)

Should be fixed. --Michael Mol 23:23, 14 December 2011 (UTC)

Thanks Michael! --Paddy3118 04:18, 15 December 2011 (UTC)