Wikipedia talk:Wikipedia Signpost/Single/2016-11-26

Comments

The following is an automatically-generated compilation of all talk pages for the Signpost issue dated 2016-11-26. For general Signpost discussion, see Wikipedia talk:Signpost.

Blog: The top fifteen winning photos from Wiki Loves Earth (356 bytes · 💬)

← Back to 2016-11-26

Discuss this story

• Beautiful. All of these photos are breathtaking. -- œ^™ 11:33, 18 December 2016 (UTC)

Featured content: Featured mix (707 bytes · 💬)

← Back to 2016-11-26

Discuss this story

• What is Taylor eating? And that's my way of saying that won't be a featured photo.— Vchimpanzee • talk • contributions • 17:39, 2 December 2016 (UTC)
  • She's a professional singer holding a wireless microphone to her mouth, as is quite appropriate and proper. Praemonitus (talk) 17:20, 5 December 2016 (UTC)

In the media: Roundup of news related to U.S. presidential election and more (2,135 bytes · 💬)

← Back to 2016-11-26

Discuss this story

• Interesting editorial in the Richmond Times-Dispatch: The fake epidemic of fake news. (Reprinted at reason.com with better formatting.) --Guy Macon (talk) 09:17, 26 November 2016 (UTC)

• And another from the Washington Times: Using fake news against opposing views. --Guy Macon (talk) 20:17, 26 November 2016 (UTC)

• Would an OTRS volunteer be kind enough to process the OTRS ticket for the Successor album mentioned in the story? Kaldari (talk) 22:00, 29 November 2016 (UTC)
• I never went to Facebook at home. The fact that a computer at my church says don't go to Facebook tells me it's not safe and I avoid sites that aren't safe at home. Plus with my slow Internet my computer can't handle the information overload. I have less than 40 friends and go to most of their timelines maybe once a week at a library. One person in particular has a ton of what I believe to be legitimate news stories. Under these circumstances, how much do you want to bet I don't actually read any of those stories? Plus if I click on anything, I don't get back the person's timeline the way it appeared before I clicked. I get a few recent posts and then have to wait ... and wait ... and wait ... for all the other stuff I already read to load once again.— Vchimpanzee • talk • contributions • 17:57, 30 November 2016 (UTC)

News and notes: Arbitration Committee elections underway (5,080 bytes · 💬)

← Back to 2016-11-26

Discuss this story

A minor correction - the article lists me as someone whose term is about to finish, but I actually left the Committee at the end of 2015. -- Euryalus (talk) 09:14, 26 November 2016 (UTC)

Thanks for pointing that out. Better now? -Pete Forsyth (talk) 09:28, 26 November 2016 (UTC)

Yes, and thanks for indulging my enthusiasm for pedantic detail. -- Euryalus (talk) 09:32, 26 November 2016 (UTC)

If there's one thing that binds us wikifolk together, that enthusiasm is probably it. -Pete Forsyth (talk) 09:38, 26 November 2016 (UTC)

More tedious nitpicks! 1) Yunshui resigned in 2015 and is not a current arbitrator. 2) I think you've mixed up Mkdw (who has never run before) and Ks0stm. 3) Salvidrim! withdrew in 2015 before voting began (not being on the ballot is one way to run unsuccessfully... ;) Opabinia regalis (talk) 09:55, 26 November 2016 (UTC)

Thanks. If Tony1 doesn't get to this in a few hours, I'll adjust. -Pete Forsyth (talk) 18:44, 26 November 2016 (UTC)

@Peteforsyth: I've made the correction regarding Ks0stm and myself. I apologize if this is a breach in protocol with the Signpost but I hope you can understand my concerns about the potential implications it might have on the election results if information about the candidates is incorrect. Mkdw^talk 19:43, 26 November 2016 (UTC)

Indeed, thank you Mkdw. -Pete Forsyth (talk) 19:50, 26 November 2016 (UTC)

Of the seven arbitrators whose terms are about to finish, four are standing at this election for another term, and four are not contesting their seats But 4 + 4 = 8. I think you mean three are standing. Hawkeye7 (talk) 09:56, 26 November 2016 (UTC)

Fixed. Graham87 10:53, 26 November 2016 (UTC)

Thank you both. -Pete Forsyth (talk) 18:44, 26 November 2016 (UTC)

• The "Brief Notes" section says this about Wikipedia Asian Month: "Taking place throughout November 2016, the purpose of Asian Month on the English Wikipedia is."

The purpose is...what, exactly? Unless I am very much mistaken, that appears to be an incomplete sentence. Joshualouie711 (talk) 14:31, 26 November 2016 (UTC)

Thanks Joshualouie711, that's a vestige from a copy-paste. The associated text is in the first sentence. Thanks for pointing itout, I'll trim it out. -Pete Forsyth (talk) 18:44, 26 November 2016 (UTC)

As a minor correction, please note that my prior ArbCom service was from 2008 to 2014 (not 2007-2014). Thank you. Newyorkbrad (talk) 22:41, 26 November 2016 (UTC)

(edit conflict) Fixed a couple more of these. Opabinia regalis (talk) 04:25, 27 November 2016 (UTC)

• Perhaps this counts as a bigger nit. "Through this formula, the minimum score (support per (support+oppose) required for election is 50 (not itself a percentage)". So, how can this quotient, using natural numbers, yield a result (not a percentage) greater than one? Jim.henderson (talk) 00:34, 27 November 2016 (UTC)

  Good point. 0.5 was intended, not 50. Corrected now, and clarified. -Pete Forsyth (talk) 04:19, 27 November 2016 (UTC)

• Readers, apologies for errors. I have so much RL work pressure at the moment that it was hard to find the space to return to the text for pre-publication checking. Tony (talk) 01:12, 29 November 2016 (UTC)
  • And from me as well -- I edited the piece, but did not double-check these important details. Thank you to all for the observations and corrections. -Pete Forsyth (talk) 16:08, 30 November 2016 (UTC)

Op-ed: Fundraising data should be more transparent (47,115 bytes · 💬)

← Back to 2016-11-26

Discuss this story

Spending

Today the WMF is spending 300 times as much (52596782 ÷ 177670 ≈ 296) as it was spending ten years ago.

Year	Total Support and Revenue	Total Expenses	Increase in Net Assets	Net Assets at year end
2003/2004 [1]	$80,129	$23,463	$56,666	$56,666
2004/2005 [2]	$379,088	$177,670	$211,418	$268,084
2005/2006 [3]	$1,508,039	$791,907	$736,132	$1,004,216
2006/2007 [4]	$2,734,909	$2,077,843	$654,066	$1,658,282
2007/2008 [5]	$5,032,981	$3,540,724	$3,519,886	$5,178,168
2008/2009 [6]	$8,658,006	$5,617,236	$3,053,599	$8,231,767
2009/2010 [7]	$17,979,312	$10,266,793	$6,310,964	$14,542,731
2010/2011 [8]	$24,785,092	$17,889,794	$9,649,413	$24,192,144
2011/2012 [9]	$38,479,665	$29,260,652	$10,736,914	$34,929,058
2012/2013 [10]	$48,635,408	$35,704,796	$10,260,066	$45,189,124
2013/2014 [11]	$52,465,287	$45,900,745	$8,285,897	$53,475,021
2014/2015 [12]	$75,797,223	$52,596,782	$24,345,277	$77,820,298

During this time the actual cost to run the servers ("you will probably be asked to donate the price of a rather expensive cup of coffee to keep our servers running" in the fundraising banner) has remained roughly the same. --Guy Macon (talk) 09:30, 26 November 2016 (UTC)

Thanks Guy. However, I'm not entirely clear on how that connects to the piece above? Could you elaborate a bit? effeietsanders 10:25, 26 November 2016 (UTC)

In my opinion, pretty much everything you wrote was spot on right up to the conclusion. We really do need more transparency in all of the areas you covered. However, your conclusion ("Let's make optimal use of the expertise that our range of volunteers has to offer in our movement for fundraising optimization, and provide our volunteer base with the tools to help our mission in the best way possible!") is the equivalent of someone on a train calling for everyone to work together to make the train run smoother and go faster while ignoring the larger picture, which is that the train just entered a partly-constructed tunnel that only goes half way through a mountain at full speed. "Fundraising optimization" is a worthy goal, and you have presented some great ideas for optimizing fundraising, but our current ever-increasing spending trends are based upon the assumption of ever-increasing donations. One day the revenue is going to crash. and at that time it will be to late to dial back the overspending. --Guy Macon (talk) 12:51, 26 November 2016 (UTC)

How the money is being spent (and whether those expenditures are necessary) is an interesting topic to discuss, but quite a different one. This post was about ways the interaction between the WMF Fundraising department and the community/affiliates could be improved so that the volunteers can work more effectively. That is independent from how much money has been raised, how it is being raised and how it is being spent. I don't quite agree with your oversimplification, but I think it would be distracting right now to go into detail on that. I'm glad to hear you do agree on my points with regards to the topic of my post. effeietsanders 14:52, 26 November 2016 (UTC)

Holy crap, I had no idea the WMF was sitting on, if I'm reading that balance sheet right, around $67 million in investments. I've glanced at the financial reports before, but all that mind-numbing accounting jargon... How does that compare to similar nonprofits? It looks like the WMF is a mutual fund that just happens to have a side business. Sure would be nice to put some of that money towards, y'know, improving the content on the projects. How many academic database subscriptions would that buy? But, I can already hear the refrain of "We're just a tech outfit! It's up to the volunteers to create all the stuff that attracts the donors who pay our salaries!" I wonder what the WMF response would be if the volunteer community decided to use their online real estate to hold their own fundraiser for improving content. (I have a guess.) --47.138.163.230 (talk) 11:17, 26 November 2016 (UTC)

There are other nonprofits with large reserves (The Harvard University endowment is at $40 billion US dollars, but they have been building it up since 1636), but I can find no nonprofit other than a few Ponzi schemes that has increased spending by %29,600 (!) over ten years while doing essentially the same amount of actual work on their main reason for existing, which in our case is keeping the servers running. --Guy Macon (talk) 12:51, 26 November 2016 (UTC)

In the UK the Charity Commission, the government regulator for non-profits, used to recommend that all charities should target having reserves equal to 2 years spending (by no means all actually have this), though new guidelines published this year have dropped that. Apparently the average level for UK charities is 15 months spend. WMF has some way to go before reaching this level. Johnbod (talk) 15:31, 26 November 2016 (UTC)

Good op-ed! Icebob99 (talk) 17:44, 26 November 2016 (UTC)

• I agree with effeiets, what ever the money is used for, we should try our best to fundraise as best as possible. And while 50+ million USD is a bit of money, I assume that much that was not done during the first years have been fixed now, like improved servers, backup etc. Part of the problem may also be that WMF is headquartered in San Francisco, possibly not the cheapest place to have an office. Maybe a major saving could be achieved by moving the WMF to Ireland? Still English-speaking, but much cheaper housing, lower vages etc. while at the same time a nice country that may easily attract talent from both the US, EU and the rest of the world. Ulflarsen (talk) 19:58, 27 November 2016 (UTC)

• It costs slightly less than $2.5 million USD per year to keep the servers running.[13] --Guy Macon (talk) 23:10, 27 November 2016 (UTC)

The cost of running the servers will remain the same no matter where the WMF offices are located: the servers are in Florida, & will likely remain there for the foreseeable future. On the other hand, I don't see any benefit to keeping the WMF offices in San Francisco. The argument was that they needed to be located there to take advantage of the skilled pool of talent, but I haven't seen that happen: the people the WMF hired locally have tended to stay a very short time, & in that time have not offered any advantages to justify remaining in SF. Almost any other location in Europe or North America would be less expensive to house the WMF headquarters, & a number of locations/cities amongst them that offer an acceptable variety of amenities for the staff. It is past time to re-evaluate having WMF in San Francisco. -- llywrch (talk) 00:33, 28 November 2016 (UTC)

One small correction: the WMF technical infrastructure is not located in Florida; it is currently built of two main datacenters in the US, two caching POPs (one on the US west coast, one in the EU) and some networking POPs [14]. GLavagetto (WMF) (talk) 10:03, 28 November 2016 (UTC)

We need to stay in the USA as our work is supported by the current laws in that country. Moving is such a pain that I imagine that is part of the inertia to relocating. I imagine the option is on the table though expecially with so many people working remotely. Doc James (talk · contribs · email) 00:53, 28 November 2016 (UTC)

Wikipedia:Avoid Parkinson's Bicycle Shed Effect / Law of triviality. The solution to this problem is not to try to identify specific places to cut costs. The solution is to elect members of the board of directors who are committed to full transparency regarding where your donations go. As an experiment to prove that we have a transparency problem I spent over a year trying every avenue anyone suggested to try to get an answer to the question "May I see an itemized list of exactly what computer equipment and office furniture was purchased with the $2,690,659 spent in 2012 and the $2,475,158 spent in 2013?". I was stonewalled. See Wikipedia:Wikipedia Signpost/2015-03-18/Op-ed#Computer equipment and office furniture for one of my many attempts. --Guy Macon (talk) 05:47, 28 November 2016 (UTC)

There is more to leaving San Francisco than saving money. Having been a member of Wikipedia since almost the earliest days (having joined October 2002), I can attest to a change in the philosophy of the organization, which I suspect is due to having the headquarters in San Francisco. WMF wants to be part of Silicon Valley, & consciously or not copies many of the habits of that group: emphasis on technology, an addiction to making every action a "world-changing" achievement, & even the lack of transparency you mention, Guy. Startups are notoriously secretive, & compulsively insist on everyone involved signing non-disclosure agreements. NDAs are a fact of life in Silicon Valley.

I would hope that if the headquarters were moved to another city not so tightly caught up in the high tech startup mindset, say Denver or Chicago, maybe the WMF could recover some of the ideals we advocated in the old days; saving money is simply a tangible argument to achieve the move, & in the worst case what we would achieve. But I wouldn't be surprised if various people with far more clout than I want the Foundation to remain where it is due for the prestige, the effects of the Silicon Valley mindset be damned. -- llywrch (talk) 07:52, 28 November 2016 (UTC)

To the above I would add that being "notoriously secretive" in the specific case of Wikipedia is just plain stupid. Do they imagine that there is some other online encyclopedia that will steal our best ideas, with "best ideas" being not our content or the way we run the encyclopedia -- both are freely available to all to copy -- but rather how every year we manage to spend $12,000 per employee on furniture on computer equipment and furniture?[15] --Guy Macon (talk) 16:16, 28 November 2016 (UTC) Edited 11:17, 4 December 2016 (UTC)

I'm old enough to have learned that just because something is "just plain stupid" doesn't mean someone (singular or plural) won't do it. Even if forewarned. -- llywrch (talk) 16:42, 28 November 2016 (UTC)

I'd like to interject here to point out that the WMF (in the person of Garfield Byrd, the former Chief of Administration) put to bed the "$12,000 on furniture" meme some time ago, saying in April of last year, "Of the amounts quoted most of the money was spent on servers and other equipment for the data center. The next biggest amount was for computer equipment and software for staff. Only about 9% of the amount in 2013 and about 7% of the amount in 2012 was spent on furniture and fixtures."

I don't disagree with my friend Lodewijk's op-ed or conclusions (indeed, while employed at WMF I made many of these same arguments repeatedly), but I would think we should strive for evidence based discussion, not relying on old and discounted memes. -Philippe (talk) 06:46, 4 December 2016 (UTC)

That's rather misleading claim. After a year and a half, I am still being stonewalled on the specific question that I have asked dozens of times on multiple pages. Are you or anyone else prepared to answer the question? If not, are you or anyone else prepared to be honest just tell me straight out that the WMF has no intention of ever answering my specific question? As a reminder, the specific unanswered question is: "Where can I go to see an itemized list of exactly what computer equipment and office furniture was purchased with the $2,690,659 spent in 2012 and the $2,475,158 spent in 2013? Verifying that those purchases were reasonable and fiscally prudent would go a long way towards giving me confidence that the rest of the money was also spent wisely. Needless to say, nobody needs to know who got what; an accounting with all personal information redacted is fine." --Guy Macon (talk) 11:17, 4 December 2016 (UTC)

---

• This is a great article. I am glad that it is in The Signpost, because now we can all take for granted that everyone in fundraising at the Wikimedia Foundation has read this and knows about the issue. The article makes a strong case for transparency. There is a conservative request for information. So far as I know, no one has published anything to counter this request or explain why it should not be fulfilled. These ideas about sharing information from fundraising have been in circulation for at least 5 years and it is great that Effeietsanders put them into print here to raise them again to urgency. If more information is not shared, then at least now the fundraising team is aware that they are supposed to give an explanation of how they decide what to share. Blue Rasberry (talk) 13:26, 28 November 2016 (UTC)

You will never see anything countering this request or explaining why it should not be fulfilled. The only response from the WMF will be stonewalling (ignore it and it will go away). I proved this with my experiment asking about furniture detailed above. --Guy Macon (talk) 16:16, 28 November 2016 (UTC)

Guy Macon The Wikimedia Foundation is addressing a series of concerns which are difficult or impossible. I will not defend any shortcomings, but to counterbalance the problems, I see processes which improve with time and discussion and that development makes me feel encouraged. We are in an environment with hundreds or thousands of major issues, and it is beyond any individual human's ability to browse them all. I do not expect that all urgent discussions must happen simultaneously and presently. Your judgement may be that not enough is improving, or that things are getting worse. For me, I see enough improving at a pace which pleases me that I can keep my faith in the process. To me, the WMF seems to be moving quickly as compared with other organizations of comparable size.

You have made your request for budget discussion repeatedly for a couple of years now. Perhaps when you post this again you can link to a list of instances in which you have raised the issue, that way you create supporting evidence that you have been seeking a response over time. I am sure other people know your name and this cause you keep raising. I want this discussed but I expect that many others are like me, and do not have time or capacity to make every issue the top priority for discussion. Just because more people do not drop what they are doing to rally behind an issue does not mean that they do not feel that it is important, or that they do not think about it. I like what you are doing. I wish it were easier to sort the most urgent causes among the many which are raised. The one you raise also seems important. I am not sure how it places among the top 100 issues. Blue Rasberry (talk) 17:22, 28 November 2016 (UTC)

Ah, the the top 100 issues. That's the Catch-22. I started all of this long ago by making a small effort to open up discussions on [A] WMF spending and [B] WMF software development. The (quite reasonable) responses I got were all about these being big, complex issues with a lot of opinions on what the solution is. I agreed with those arguments, so I carefully designed the following two very small, very specific, and rather noncontroversial suggestions:

• Accounting: "May I please see an itemized list of exactly what computer equipment and office furniture was purchased with the $2,690,659 spent in 2012 and the $2,475,158 spent in 2013? Why am I unable to get anyone at the WMF to answer this question, even if the answer is "no"?" [16]
• Software development: "In the HTML that Wikipedia sends to the browser, every line sent to the browser has a DOS-style carriage return and line feed (OD OA) as a line ending. If we used a Unix-style line feed (0A) that would save one byte on every single line of HTML on Wikipedia. Actually HTML works just fine with both the carriage return and the line feed removed, but let's just discuss (OD OA) vs (OA) for now. We could do this by making the line ending configurable in the preferences with the default (OD OA) and (OA) an option, then after we are sure there are no bad effects, change the default. Why am I unable to get anyone at the WMF who has the ability to make this change to discuss the merits of doing this?" [17] [18][19]

In both cases I was stonewalled, and I fully expect the result of this discussion to be the same.

So if I propose a big change, it is shot down for reasons that even I agree with, and if I propose a small change, it is stonewalled because it isn't a big change. Thus no change I propose will ever be evaluated. Catch-22. --Guy Macon (talk) 18:10, 28 November 2016 (UTC)

I agree we need greater transparency around financials. We should be transparent by default and only hold stuff back when specific justification is present. The justification used should be provided when details are held back. Doc James (talk · contribs · email) 18:20, 28 November 2016 (UTC)

---

I have been asked to do an editorial for the Signpost concerning my views expressed above, and I am doing research for it. Specifically, in 2005 Jimmy Wales told a TED audience the following:

"So, we’re doing around 1.4 billion page views monthly. So, it’s really gotten to be a huge thing. And everything is managed by the volunteers and the total monthly cost for our bandwidth is about 5,000 dollars, and that’s essentially our main cost. We could actually do without the employee … We actually hired Brion because he was working part-time for two years and full-time at Wikipedia so we actually hired him so he could get a life and go to the movies sometimes."

First question: how many page views per month are we seeing ten years later (any figure from 2015 or 2016 will do)?

Second question: is there any reason to believe that bandwidth costs per page view has gone way up or way down in the last ten years? --Guy Macon (talk) 01:25, 29 November 2016 (UTC)

We are seeing about 16 billion pageviews a month across all projects and all languages.[20]

Are bandwidth / data hosting costs more per pageview? I imagine yes as we now have two data centers in the US and are looking at maybe one in Asia. I would imagine our up times today are better than they were in the past.

How fast we serve pages is likely also much better but still varies globally. While Europe and North America see fast load times much of the rest of the world does not so would be good to see more money spent their.

The WMF does much more than just serve pages. I see the development of our apps and our mobile page as being exceedingly important as more than half of medical pageviews are now read on mobile.[21]

The visual editor (even though I still do not use it personally) is much loved and nearly exclusively used by the students I teach. Wikipedia zero gets about 70 million pageviews a month and reaches a demographic that would be hard to reach otherwise. The existence of this program also provides excellent leverage to convince organizations like WHO and ECGepedia to work with us. Etc :-) Doc James (talk · contribs · email) 05:45, 29 November 2016 (UTC)

Other work that has been key has been that of the community tech team. They have helped further developed the copypatrol tool which not only allows us to keep a better handle on copy and pasting but paid editing.

The content translation tool has been useful for developing medical content in further languaging and bringing further non Wikipedia translators into the fold. We have introduced many to Wikipedia with that tool.

I am happy to see the WMF involved in more than just servicing pages. Doc James (talk · contribs · email) 05:57, 29 November 2016 (UTC)

Many people have their favorite parts of the WMF's ever-increasing spending. You like visual editor, others think it is a disaster. I am sure that someone out there will defend the twelve thousand dollars per employee per year spent on furniture or the ten thousand dollars per employee per year spent on travel. We could even find someone who defends having a 2000-person wikimania in a mountainous village with a population of 750 in northern Italy. The problem is that the sum total of your desire to spend money plus all of the other people who desire to spend money is fueling the WMF's ever-increasing spending and putting us on the road to ruin. --Guy Macon (talk) 11:15, 29 November 2016 (UTC)

We are not spending money that we do not have so not sure how this is going to ruin us? From what I understand the plan is for expenditures and fundraising goals to level off at around were they are currently.

With respect to visual editor, I know of many new editors who like it. As I said I do not use it myself. Doc James (talk · contribs · email) 18:41, 29 November 2016 (UTC)

Re: "From what I understand the plan is for expenditures and fundraising goals to level off at around were they are currently",^{[Citation Needed]}. I have been closely monitoring the many places where the WMF discusses finances, and have seen nothing even hinting at this -- but of course I may have missed something. Re "We are not spending money that we do not have so not sure how this is going to ruin us?", as a physician I am sure that you are aware that in its early stages cancer does not as a rule consume resources that the body doesn't have, and indeed that the real problem is that the cancer just keeps growing and growing, and that it is this ever-increasing growth that promises ruin for the patient, not the current effects of the small malignant growth. --Guy Macon (talk) 21:32, 29 November 2016 (UTC)

@Guy Macon: Substituting a carriage return and line feed with a line feed is unlikely to have any effect on page weight. All of our pages are served with gzip compression, so common character sequences are substituted during transmission anyway. The WMF has actually done a lot of impactful work on page weight and rendering speed in the past couple years.[22][23][24][25] Kaldari (talk) 22:16, 29 November 2016 (UTC)

Nevermind, I see that you were already informed about the gzip compression over a year ago. Strange that you still claim no one at the WMF will discuss it with you. Kaldari (talk) 22:28, 29 November 2016 (UTC)

My question was not "Why am I unable to get anyone at the WMF to discuss the merits of doing this?" It was "Why am I unable to get anyone at the WMF who has the ability to make this change to discuss the merits of doing this?" A developer working on page weight could do a quick test in less than five minutes that will answer the gzip question using the actual Wikipedia environment. Instead, I am being asked by people who have zero ability to actually make the change to duplicate Wikipedia and do my own tests, and of course if I do that I will be then be told that I have not duplicated things like the caching that a large site with multiple servers uses. Plus, due to the overhead and latency of compression and decompression, you should only gzip files above a certain size threshold (often chosen to be between 150 and 1000 bytes), so I would also have to figure out the size at which Wikipedia stops compressing, estimate how many pages are below that (redirects are tiny), and in the end I still will have utterly failed to open up a discussion with an actual developer who has at least the potential of running a real-world test of my proposal. So yes, I gave up. I see a lot of work with zero chance of success. --Guy Macon (talk) 23:10, 29 November 2016 (UTC)

@Guy Macon: FYI, the person who replied to you on Phabricator is a Principal Software Engineer at the WMF. They would definitely have the ability to make the change you are asking for. However, I don't think it's likely that anyone is going to pursue this idea. If MediaWiki wasn't using gzip it would be a great idea, but as it stands it just isn't likely to have any practical effect. Kaldari (talk) 08:35, 12 December 2016 (UTC)

I stand corrected. If he indicated that he was a developer or offered to be helpful I missed it. You have no idea whether the existence of gzip invalidates my idea. It is just a guess on your part. Many wikipedia pages are not gzipped. We don't know how many or how much traffic they get. Some browsers don't support gzip, and not supporting gzip may be more common in the third world countries that my proposal addresses. Finally, compression is not free. If Wikipedia's gzip no longer has to compress carriage return/line feed sequences, that frees up a slot in the uses-few-bits part of the token space for another sequence, and all the other sequences move up into slots that require fewer bits. What is the end result? Very little gain or substantial gains? I don't know and have no way of finding out. A Principal Software Engineer at the WMF could run a couple of experiments in the real Wikipedia server environment in well under half a day and tell us whether there is enough space savings to make this feasible.

Please keep in mind that after the WMF stonewalled me (stonewalling: not evaluating and thus never actually accepting or rejecting a suggestion) on my proposal (Start a project -- a real project with measurable goals and a schedule -- to reduce page weight), I picked the smallest and easiest to implement suggestion that I could think of to see if they would stonewall me on that one as well. The result has been more than sufficient to cause me to abandon both suggestions and to never again make any suggestion about WMF software development. I quit, but I will never agree that I wasn't stonewalled. --Guy Macon (talk) 16:59, 13 December 2016 (UTC)

Editor butts in: All, I'm pleased to see so much interest in developing ideas for a new op-ed. However, one of the reasons I initially suggested an op-ed is that Guy Macon's main point is different from that of effeietsanders, and worthy of a discussion space of its own. Possible to move this discussion somewhere else? I'd suggest creating a draft (even if it is only a rough sketch, early on) by clicking the "Create new op-ed draft" button here. And to the extent there are diverging views, I'd be happy to consider more than one piece, from different authors; that way, there's no need to work out fundamental disagreements prior to publication.

Any further comments on transparency or collaboration with other organizations? -Pete Forsyth (talk) 23:49, 29 November 2016 (UTC)

• Every time I go to Wikipedia at a library I am subjected to this fund-raising banner and I can't get rid of it. I know people will tell me to sign in, but when that banner is there, there is no place to do so. I want to sign in anyway because I don't want my edits recorded by IP address and my contibutions help me to decide what to do while I'm there.— Vchimpanzee • talk • contributions • 18:06, 30 November 2016 (UTC)

Hello all,

This op-ed almost relates to many of the responsibilities that fall under my role and so it was only proper that I be the one respond. With being at a partnerships and funding workshop at Wikimedia Deutschland with a number of other affiliates earlier this week, as well as the launch of our English Fundraiser on Tuesday, I have been trying to find the time so that I could properly respond.

I wanted to acknowledge the issues raised by Lodewijk regarding country-level data. I can’t give any further background or information regarding this decision, but I do genuinely recognise the frustrations with that. As someone working in community engagement, and someone who has been involved in the community for the last ten years, one of the most difficult things to do is to give an answer that you know doesn’t satisfy requests for information.

In terms of the work that my department does, trying to involve more community involvement in the fundraiser has become a major goal over the last year. You’ve already noted that for the last two years we have had an open page calling for fundraising ideas, which produced many suggestions whether in terms of banners or processes that actually got implemented. In fact my very role came about partially from that very page. It was a direct recognition that we needed to do better in terms of engaging the community in the department's activities.

This year we have been trying to collaborate more closely with local affiliates regarding translations, localization and the type of messages we use in our fundraising appeals and communications, in order to make them effective, culturally appropriate and relevant to the specific community we fundraise in. This year we worked closely with staff at the Swedish and Dutch chapters, brainstorming and editing copy for our fundraiser. We experimented with an annual email newsletter in conjunction with Wikimedia Italia and we supported Wikimedia Hungary, Italia and Poland with their annual tax campaigns to make them more effective. These are small steps but we are slowly trying to build on these relationships, treating them as the foundations for much cooperation across the movement. Reviewing translations should only be a minimum, and there is genuine aim for the community's efforts to be a lot more than that.

There are definite challenges when it comes to scheduling of campaigns. Some chapters plan their CentralNotice activities well in advance as part of their annual planning, other chapters take a more flexible approach where they go with opportunities that arise through the year and communities that are more fluid or come together for specific event are more reactive and their planning occurs much closer to events they run. Likewise we put together rough timeline for campaigns for the following financial year (July-June following year) that is normally put in place in the two months prior to the start of that financial year (May & June). The further we plan the more uncertainty contained within those plans. There are changing dependencies stemming from technical work, work with payment processors sometimes simply changing campaign priorities which can include unexpected changes in previous campaigns. The known larger campaigns we do our best to work around (WLE, WLM & CEE) along with regular local campaigns (WikiFranca). The aim is to reach out to chapters at least 4 months in advance and in some instances up to 6 months. We published our timeline for the first 8 months of the year on the CentralNotice calendar which was as far as we felt comfortable in our plans, and we are trying to firm up the plans for the remainder of this financial year and I’ve begun reaching out to affiliates where we are planning on doing online fundraising in their geography. We haven’t had a major conflict in 13 months and I hope that the “Days since last accident” number continues to grow in value.

So in short, yes, there is a lot that can be done to improve the way we work with the community but actively trying to build that stronger working relationship with our affiliates and the wider online communities. Seddon (WMF) (talk) 04:51, 2 December 2016 (UTC)

@Seddon (WMF): The author states "It is plainly embarrassing for volunteers, advocating for transparent and openly licensed information flows, to say they don't even know remotely how much their movement collects in contributions from their own country." To simply acknowledge that the WMF is deliberately withholding this information and you understand there's frustration is insufficient. It is foolish to ask volunteers to help WMF better exploit their countrymen with ad campaigns but to then leave those same volunteers holding the bag. Chris Troutman (talk) 18:03, 13 December 2016 (UTC)

MOS BOLD

I'm glad to hear Effeietsanders's opinion, especially coming from someone outside en-wp. I am offended, however, by the use of bold in the body text. The abuse of boldface, italics, and underlines are well-known in the fundraising world as methods to snag the drive-by reader; to make them hear the pleas for help. I can't help but assume the writer thinks I am cattle based on their effort to moo and bray at me. Please do not continue to condescend. Chris Troutman (talk) 19:19, 30 November 2016 (UTC)

To Chris troutman and anyone else interested in the Signpost's quality: Please do get in touch if you'd like to volunteer. We will make a more comprehensive statement about the best ways to engage in the coming weeks; but in the meantime, as always, I am committed to finding a good fit for anyone who has a sincere desire to help. (This editorial note from October 2015 is still a fairly accurate overview of how we work.) If copy editing is your thing, I'd especially suggest getting in touch with Montanabw -- though I am also happy to discuss, and our HR person Rosiestep is also a good resource. -Pete Forsyth (talk) 02:30, 1 December 2016 (UTC)

@Chris troutman: I'm sorry to hear this use of layout offends you. You're correct that I'm not a typical member of the English Wikipedia community. I guess the use of layout is a matter of taste: I like to use boldface in pieces like this (opinionating articles, not encyclopedic articles), to indicate clearly to the reader which key phrases are most important to me. I consider this to be especially helpful to the non-native speakers. Boldfaced text would, in my texts, typically be used to indicate importance, while italics would be used to indicate quotations and/or foreign phrases. Why do you feel that the use of such text to indicate importance would be condescending, given the aim at a wide audience? effeietsanders 15:58, 1 December 2016 (UTC)

@Effeietsanders: I enjoyed your piece. The content you communicated was important. Perhaps you felt you had to draw attention to certain phrases while aiming for "a wide audience". But as I mention, this abuse of font to draw in the reader is found in advertising pitches, especially for any number of non-profit groups like UNICEF or Habitat for Humanity in their letters begging for donations. I'm not the lowest common denominator and that's my objection. It would be akin to going to see a hilarious stand-up comedian only to see them put on clown shoes and a big red nose in an effort to get some members of the audience to laugh. I encourage your continued participation on en-wp, as I'd like to know that the editor frustration we feel here is not exclusive to Anglophones. Concurrently, I'm a customer for your content and cheap efforts to get my attention will only leave me unhappy. Chris Troutman (talk) 22:24, 1 December 2016 (UTC)

Personally, I LOVE CHEAP EFFORTS. I would be very offended if they are not used MoRe OFTEN. Who let this coulrophobic Chris Troutman comment on here? He uses SPACES in his text. The gawdy departure from scriptio continua is really galling to many. I am extremely unhappy and would like a refund for this issue.--Milowent • ^hasspoken 01:20, 3 December 2016 (UTC)

(e.c.) Chris, I agree with your objection; there's good reason that MOS discourages the speckling of bold face (it's not "font") in main text. It was fixed, and that should have been the end of it. Lodewijk has written a good, though-provoking piece, so let's concentrate on the issues he raises. Tony (talk)

Endowment

Better transparency is a good goal in both fundraising and spending, and starting this year, for the Wikipedia Endowment. Is there more information available regarding the endowment than this blog post or this Wikipedia15 article? Was there a discussion on Wikipedia or Wikimedia about which company would be managing the endowment, or do editors not have influence on this kind of decision? Community involvement might be beneficial. For example, one concern might be whether management fees are too high. The Tides website has a one-page summary of the collective action funds. The summary describes an unusually high management fee of "3-5%." Generally, management fees don't include fund fees, which might add an additional 0.2 to 3%. There are many concerns that could be ameliorated with greater transparency, such as endowment volatility, diversification, impartiality and influence, ethical investing, along with many other issues. This is likely a can of worms, so an article discussing the matter would be helpful. -kslays (talk • contribs) 02:20, 16 December 2016 (UTC)

That appears to be good news indeed, but of course the devil is in the details. Assume for the sake of argument that contributions crash to less than 10% of the current numbers. Further assume that the WMF keeps spending, rapidly burning through the reserve. Would the WMF then be able to withdraw funds from principle of the endowment to continue spending? Would the endowment be at risk if the WMF went bankrupt? --Guy Macon (talk) 09:25, 16 December 2016 (UTC)

(...Sound of crickets...) --Guy Macon (talk) 11:58, 18 December 2016 (UTC)

No one is going to respond to you, man. You don't accept the answers you're given, (redacted). Why are you surprised that you get no response?--Jorm (talk) 21:45, 18 December 2016 (UTC)

Please note that Jorm has an unrelated problem with me that may be influencing his behavior. We had a previous disagreement about the quality of some software he worked on when he was with the WMF. --Guy Macon (talk) 22:56, 15 August 2019 (UTC)

Need help updating image for Wikimedia financials

At User:Guy Macon/Wikipedia has Cancer the table has been updated for 2017-2018 but the image at commons:File:Wikimedia Foundation financial development multilanguage.svg only goes to 2016-2017. Could someone with SVG editing skills please look at that table and update the image? --Guy Macon (talk) 22:56, 15 August 2019 (UTC)

UPDATE: I solved the problem on my page by replacing the image with a template. The other pages that use the image still need an updated version. --Guy Macon (talk) 02:01, 16 August 2019 (UTC)

Special report: Taking stock of the Good Article backlog (10,995 bytes · 💬)

← Back to 2016-11-26

Discuss this story

The results of gamification

The assertion that "We now know that the GA Cup does not lead to "drive-by" passes" has no basis in fact. It may be true that insufficient reviews occur at the same rate so the Cup doesn't encourage the practice but let's remember that the number of bad reviews is increasing at the same time reviews, generally, are increasing. Doing GA reviews sucks because it's actual work; I have more fun doing GOCE drives. I think the Good Article WikiProject is key to objectively improving content whereas GOCE is by-and-large just fixing word salad, which almost anyone can do. Efforts like the GA Cup are our collective means of putting these articles to stringent standards. GA status is often, though not always, a precursor to pursuit of A-class or FA. I remain concerned that these contests (of which I am a part currently) attract editors who are still unfamiliar with proper reviewing. It's demoralizing to see bad reviews done, especially when you're competing for points. WikiProject Articles for Creation had 8 drives since 2012. The last drive saw a lot of poorly done draft reviews and the results were so skewed that the WikiProject hasn't held another drive since 2014. I would hate to have these drives ruined by bad editing and we can only rely on the judges of the competition to stay alert to malfeasance. Chris Troutman (talk) 21:17, 26 November 2016 (UTC)

I'm confused by your point. If you're saying judges should stay vigilant of quick passes, I agree. That is not exclusive of the fact that there's no evidence to support a claim that quick passes are increasing, because indeed my argument is based in fact. To quote the paragraph prior to your quotation: Comparing five months before with the four months during the first GA Cup, there is no significant difference between the pass rates during or before the GA Cup ( t(504.97)=-1.788, p=0.07 ). In fact, may have actually decreased slightly, from 85% beforehand to 82% during the cup and because the p-value is close to significance. If more drive-by passes were occurring and more reviews were happening then we would see a higher rate of passage during the GA Cup. There is no significant difference. So either there was no change in the number of reviews or there was no change in the rate of drive-by passes. There clearly is an increase in the number of reviews, that's why the backlog decreased, so the only remaining reason for the result is that there was no change in the rate of passes.

This, admittedly, is an operational definition that doesn't get fully at the answer. It assumes that there was not a substantial amount of quick-passes outside of the Cup and that whatever the ratio of quick-pass to non-quick-pass was outside of the cup is the same as during the cup. You can dispute these assumptions, but based on all the data I have available to me there is no evidence that the GA Cup causes drive by passes. That claim is far from having no basis in fact, and is far more factual than anecdotal gripes. Wugapodes [t^hɔk] [ˈkan.ˌʧɻɪbz] 04:00, 27 November 2016 (UTC)

I don't really buy into the "gamification" of this (and various similar "challenges", "drives", etc.). Maybe it really does motivate a few people, but not everyone feels competitive about this stuff. The very nature of GA, FA, DYK, ITN, etc., as "merit badges" for editors to "earn", and the drama surrounding that, led to a rancorous ArbCom case recently, and cliquish behavior at FAC has generated further pointless psychodramatics. We really need to focus on the content and improving it for readers, not on the internal wikipolitics of labels, badges, and acceptance into politicized editorial camps.

It might be more practical and productive to have a 100-point (or whatever) scale and grade articles on it to a fixed and extensive set of criteria, with FA, GA, A-class, B, C, Start, and Stub all assigned as objectively as possible based on level of compliance with these criteria (and resolving the tension of exactly what A-Class is in this scheme, which seems to vary from "below GA" to "between GA and FA" to "FA+" to "totally unrelated to GA or FA"). There are a quite a number of GA, A and probably even FA quality articles that have no such assessments, because their principal editors just don't care about (or actively don't care for) the politics and entrenched personality conflicts of our article assessment processes as they presently stand. I, for one, will probably never attempt to promote an article to FA myself directly, because of the poisonous atmosphere at FAC (which is now an order of magnitude worse than it was when I first came to that conclusion several years ago). I guess the good news is I'll have more time for GA work. :-) The more that FA, and some of the more rigid and too-few-participants A-class processes, start to work like GA historically has, the better. If, as Kaldari suggests below, the opposite is happening, with GA sliding toward FA-style "our way or the highway" insularity, then you can expect negative results and declining participation.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:48, 2 December 2016 (UTC)

@SMcCandlish: I like the theory of a 100-point scale, but I'm not sure how it would cope any better than the current system with the inherent uncertainty of what a "complete" article would look like, you need to know what the end result is before you can start having a percentage of it. For instance, the Tower of London is an FA, reflecting the huge amount of material that's been published about it, whereas its "sisters" Baynard's Castle and Montfichet's Castle no longer exist and in the latter case even the exact location is not entirely certain. I took both of them up to a pretty decent standard back in 2010, bringing together more information on them than was available in any one place on the web at the time and probably anywhere bar the Museum of London library, but they're still tiny compared to the Tower of London article. As it happens the former was GAN'd by someone else in 2013, and passed with minor copyedits; the latter just needs some minor work on the lede and formatting - that reminds, me, I need to dig out some photos I took ages ago... I wasn't that bothered about taking them through the GA process, and certainly have no interest in taking them to FA. Actually I think the view of GA as "a precursor to pursuit of A-class or FA" is part of the problem, to my mind we need a lot more emphasis on the good as opposed to perfection. After all, under the guidelines of many projects even a GA article is "Useful to nearly all readers, with no obvious problems; approaching (but not equalling) the quality of a professional encyclopedia", but one FA takes as much time as ??3-5?? GAs? In fact I'd argue the real focus should be more on avoiding bad articles than polishing the already pretty good ones. I've a little mini-project on the go where I've started on getting all the Category:Towns in Kent to a decentish minimum standard, that is still nowhere near GA but at least avoids the real horrors - my working definition is restructuring them with all the sections of WP:UKTOWNS and some text and a reference in each section, plus linking in any nearby articles. So going from say this to this. Not perfect, but it's gone from an incoherent mess to somewhere in the right direction. Perhaps we could encourage people to work on the weakest articles in a set by extending the idea of GA/FA topics to C-class and B-class topics?

Age of nominations

This is very interesting read! One thing that I was looking for here that didn't get discussed was the effect of the Cup on the age of nominations - are reviews now sitting in the queue for less time than they were before these competitions started? (For those who don't know, I'm a judge in the Cup, after having competed in it the first year.)--3family6 (Talk to me | See what I have done) 05:13, 27 November 2016 (UTC)

Reviewer burnout

I took part in the first GA cup. It was a new idea with a good purpose and I felt I wasn't pulling my weight by putting more GA nominations on the pile than reducing the backlog by reviewing. Towards the end of the cup, I got burned out and reduced my activity; I still do the odd review but not as many as I used to. I know some other GA stalwarts have also stopped reviews. How can we reach out to these people and get them to participate in reviews again? Ritchie333 ^{(talk) (cont)} 14:59, 27 November 2016 (UTC)

I quit doing GA reviews when it became a tedious regimented process. I'm glad we have high standards for quality, but I miss the relatively informal process that we used in the old days. Kaldari (talk) 21:42, 29 November 2016 (UTC)

Great article

I just wanted to say that this was a really interesting read. As someone who wasn't around for the early days of the project I'd love to learn more about how some of the other now well-established processes came to be. Sam Walton (talk) 16:13, 28 November 2016 (UTC)

I second that. Although I was an active editor back in 2006 I wasn't at all involved in the GA process so it's interesting to read a succinct history. WaggersTALK 08:47, 29 November 2016 (UTC)

Traffic report: President-elect Trump (2,632 bytes · 💬)

← Back to 2016-11-26

Discuss this story

• "Trump will be the first U.S. president not to hold previous elected office or military command." Not true: Herbert Hoover was Secretary of Commerce, an appointed position, but had no prior elected or military positions. Trump, however, will be the first president with no previous governmental or military position. (Interesting aside: apparently William Howard Taft's only prior elected position was a local judgeship over two decades before he became president.) Antony–22 (^talk⁄_contribs) 04:00, 30 November 2016 (UTC)

• Antony–22, thank for that fine clarification, I have updated the entry. I should have remembered Hoover! I recently read One Summer: America, 1927 and learned more than I ever knew about him.--Milowent • ^hasspoken 04:44, 1 December 2016 (UTC)

Herbert Hoover was an interesting fellow - he gained fame with the Commission for Relief in Belgium and humanitarian efforts during the Soviet famine of 1921. As Secretary of Commerce he was credited with the "successes" in the response to the Mississippi Flood of 1927. "Successes" which did not help local African-Americans, beginning their drift from the Republican party. Those floods were also the inspiration for for the song When the Levee Breaks. Taft had important positions in the Philippines and as Secretary of War before becoming pres.--Bellerophon5685 (talk) 22:21, 2 December 2016 (UTC)

Knowing that Donald Trump & Herbert Hoover have things in common is not comforting. -- llywrch (talk) 05:18, 5 December 2016 (UTC)

Herbert Hoover and William Howard Taft proved themselves to be a competent administrators at least once before becoming President. I don't think Trumps record as head the Trump Organization is anything to celebrate ergo Trump Airlines, New Jersey Generals, Trump: The Game, Trump Steaks...--Bellerophon5685 (talk) 19:57, 5 December 2016 (UTC)