Open Source Diva named CTO Options

[Somey]

If I get a chance to ask some of the WMF leadership about this, I'd like to link to this conversation.

Don't give them any ideas!

It isn't as unusual as one might think for a company to have little or no effective disaster-recovery plan in place. CEO's are often given the choice between expensive backup/mirroring/failover solutions on the one hand, and larger executive bonuses on the other, and it's not hard to guess which they choose. Anyway, I've seen companies with all sorts of critical data on MSSQL and Oracle DB servers that didn't even have simple daily/weekly maintenance plans that could easily be set up by any intern in under 15 minutes, which apparently is also the same amount of time it takes to get a better deal on car insurance. And yet everyone has car insurance.

[Gruntled]

Why do you assume it isn't?

Because I find it amazing that to my knowledge, nobody's sued WMF in the Dutch courts by now if it's possible. For example, the copying of reproductions from the National Portrait Gallery would not be allowed under European copyright law, enforceable in Dutch courts. If you know of a Dutch case, I'd love to hear details.

This post has been edited by Gruntled: Fri 11th March 2011, 12:16pm

[Kelly Martin]

It isn't as unusual as one might think for a company to have little or no effective disaster-recovery plan in place. CEO's are often given the choice between expensive backup/mirroring/failover solutions on the one hand, and larger executive bonuses on the other, and it's not hard to guess which they choose.

I've seen prominent Wikimedians comment that they're not that concerned; they figure that if Wikimedia ever does experience a major disaster, people will come forward with mirrors and archives and quickly repopulate any lost content. In a very real sense, they've elected to crowdsource their disaster recovery plan.

[thekohser]

I've seen prominent Wikimedians comment that they're not that concerned; they figure that if Wikimedia ever does experience a major disaster, people will come forward with mirrors and archives and quickly repopulate any lost content. In a very real sense, they've elected to crowdsource their disaster recovery plan.

I will be the first to step forward with the lost article about SEO 2.0, and various others of my clients' deliverables.

[anthony]

Because I find it amazing that to my knowledge, nobody's sued WMF in the Dutch courts by now if it's possible. For example, the copying of reproductions from the National Portrait Gallery would not be allowed under European copyright law, enforceable in Dutch courts.

I think it's quite a stretch from "National Portrait Gallery hasn't sued WMF" to "Wikipedia isn't subject to Dutch law on copyright".

[Cedric]

Why do you assume it isn't?

Because I find it amazing that to my knowledge, nobody's sued WMF in the Dutch courts by now if it's possible. For example, the copying of reproductions from the National Portrait Gallery would not be allowed under European copyright law, enforceable in Dutch courts. If you know of a Dutch case, I'd love to hear details.

That thought had occurred to me also.

[Gruntled]

Because I find it amazing that to my knowledge, nobody's sued WMF in the Dutch courts by now if it's possible. For example, the copying of reproductions from the National Portrait Gallery would not be allowed under European copyright law, enforceable in Dutch courts.

I think it's quite a stretch from "National Portrait Gallery hasn't sued WMF" to "Wikipedia isn't subject to Dutch law on copyright".

I did say For example; it was the first that came to mind. And there are all those litigious individuals like Jeff Merkey who would surely find life far easier in the Dutch courts than in US ones.

[anthony]

Because I find it amazing that to my knowledge, nobody's sued WMF in the Dutch courts by now if it's possible. For example, the copying of reproductions from the National Portrait Gallery would not be allowed under European copyright law, enforceable in Dutch courts.

I think it's quite a stretch from "National Portrait Gallery hasn't sued WMF" to "Wikipedia isn't subject to Dutch law on copyright".

I did say For example; it was the first that came to mind. And there are all those litigious individuals like Jeff Merkey who would surely find life far easier in the Dutch courts than in US ones.

So, basically, you know of several people who would want to do it, and none of them have done it, so you assume it's impossible?

I'm not even sure what you're assuming. First your comment is that "Wikipedia [isn't] subject to Dutch law", which is pretty absurd. Then you say "nobody's sued WMF in the Dutch courts", implying that doing so is impossible, which is even more absurd. The two assumptions, while both absurd, are not at all the same. A lawsuit in the US courts can sometimes be decided based on Dutch law, and a lawsuit in the Dutch courts can (I believe) sometimes be decided based on US law. It's unclear which, if either, you are claiming to be "impossible" for a case involving the WMF, in addition to the Dutch courts deciding a case involving the WMF under Dutch law (which also is quite possible - presumably a dispute between WMF and Kennisnet over the colocation terms would take place in Dutch courts under Dutch law).

I'm no expert on international copyright and international libel law. With that said, I don't think the mere fact that WMF has a server located in Amsterdam means that a US citizen living in the US and with no particular ties to Amsterdam, can successfully sue a US company for libel, in Amsterdam , applying Dutch laws. I also don't think that the mere fact that WMF has a server located in Amsterdam means that a British company can successfully sue, in Dutch court, under Dutch laws, a US company for allowing a US citizen to upload a public domain image which they obtained in violation of a terms of service which no doubt said it was to be governed under the laws of a country other than Amsterdam (probably the UK). But as I said, I'm not expert, so maybe I'm wrong. (Then again, I have no evidence that Merkey is an expert on international copyright and/or international libel law and/or that National Portrait Gallery has a salaried staff expert on international copyright and/or international libel law, so the fact that they chose not to engage in such a case means very little to me.)

EDIT: In any case, maybe this is what you were trying to say, in which case I believe you are correct: "The mere fact that Wikipedia has servers in Amsterdam does not imply that any and every case against the WMF could be decided by Dutch courts under Dutch law, if only the forum shopping litigant chooses for it to be so. There are generally other factors in a case beyond where some of the servers are located."

This post has been edited by anthony: Sun 13th March 2011, 2:48pm

[Gruntled]

So, basically, you know of several people who would want to do it, and none of them have done it, so you assume it's impossible?

No, I'm asking if it's possible and expressing amazement that it hasn't been done if it is possible.

[anthony]

So, basically, you know of several people who would want to do it, and none of them have done it, so you assume it's impossible?

No, I'm asking if it's possible and expressing amazement that it hasn't been done if it is possible.

Ah, my bad. Sorry.

[thekohser]

Anyway, it's always best to give people a chance to show what they're about before assuming the worst. I say we give her at least a couple of weeks before concluding that her tenure as CTO is a complete failure.

How about 70 weeks?

[Zoloft]

... She's had a long tech career, appears to be an Open Source evangelist, coder, and advocate for women in IT.

Seems like an excellent choice, really.

I would hire her if i wanted to energize a team.

So naturally she was doomed from the start.

[Somey]

I doubt we'll ever know for certain, though Ms. Cooper could conceivably start blogging again (she stopped just before joining the WMF) and drop a hint or two. If I had to guess, I'd say it was her idea, not theirs - she probably got frustrated with being, for all intents and purposes, a glorified bandwidth optimizer and patch installer (applyer?), which wouldn't have been what she signed up for.

What she didn't stop doing (after being hired as the WMF CTO) was tweeting. At least half of her Twitter feed is about bandwidth and server problems being fixed, another 30 percent is her announcing her arrival at various airports, and the rest is just links to various OSS articles that (mostly) have little relevance to WP.

It was broadly hinted at here on WR 70 weeks ago (by K. Martin, myself, and others) that if her job was mostly going to consist of sticking band-aids on wounded servers, she wasn't a good choice for it and she wasn't going to be happy doing it. I'd say the victim here, if there is one, is probably Cooper herself. I wouldn't be the least bit surprised to find out that her job interview with the WMF board consisted of them reassuring her that she wasn't going to have to do any of the down-and-dirty stuff, and that she'd instead be concentrating on "high-level strategic technology planning" and various other things that the WMF has never cared about one iota.

If they were realistic about the nature of their organization, the WMF would give the CTO job to Brion Vibber and give him a nice big raise along with it. But they never have been, and in any event, Erik Moeller probably wouldn't stand for it.

[melloden]

I doubt we'll ever know for certain, though Ms. Cooper could conceivably start blogging again (she stopped just before joining the WMF) and drop a hint or two. If I had to guess, I'd say it was her idea, not theirs - she probably got frustrated with being, for all intents and purposes, a glorified bandwidth optimizer and patch installer (applyer?), which wouldn't have been what she signed up for.

What she didn't stop doing (after being hired as the WMF CTO) was tweeting. At least half of her Twitter feed is about bandwidth and server problems being fixed, another 30 percent is her announcing her arrival at various airports, and the rest is just links to various OSS articles that (mostly) have little relevance to WP.

It was broadly hinted at here on WR 70 weeks ago (by K. Martin, myself, and others) that if her job was mostly going to consist of sticking band-aids on wounded servers, she wasn't a good choice for it and she wasn't going to be happy doing it. I'd say the victim here, if there is one, is probably Cooper herself. I wouldn't be the least bit surprised to find out that her job interview with the WMF board consisted of them reassuring her that she wasn't going to have to do any of the down-and-dirty stuff, and that she'd instead be concentrating on "high-level strategic technology planning" and various other things that the WMF has never cared about one iota.

If they were realistic about the nature of their organization, the WMF would give the CTO job to Brion Vibber and give him a nice big raise along with it. But they never have been, and in any event, Erik Moeller probably wouldn't stand for it.

And unfortunately, this is one reason why Wikipedia hasn't really exploited its technical potential yet--it's being directed by non-tech people. People like Brion Vibber, who know MediaWiki in-and-out, should be the ones making the technical decisions--not Sue or Erik, to say the least. Sue is nothing more than a ruthless businesswoman--but when she dictates the WMF's tech directions, it usually ends up as nothing very special. The CTO could be responsible for so much more, given that the WMF has relatively few employees anyway but they have an enormous collection of tech and server stuff.

[Kelly Martin]

And unfortunately, this is one reason why Wikipedia hasn't really exploited its technical potential yet--it's being directed by non-tech people. People like Brion Vibber, who know MediaWiki in-and-out, should be the ones making the technical decisions--not Sue or Erik, to say the least. Sue is nothing more than a ruthless businesswoman--but when she dictates the WMF's tech directions, it usually ends up as nothing very special. The CTO could be responsible for so much more, given that the WMF has relatively few employees anyway but they have an enormous collection of tech and server stuff.

Spoken by someone who clearly knows absolutely nothing about technology in business. You don't want your techies making decisions about how to use technology in your business, because they'll just pick whatever technology they think is the coolest, whether or not it's the most efficient way to facilitate your business processes. The role of the CTO is to find efficient ways to facilitate business processes with technology. Brion is demonstrably unable to fill this role; as are both Sue and Erik, but all of them for different reasons.

What I'm wondering about is what Brion will do now that Danese is gone and he's back to reporting to Erik. Will he leave again?

[Somey]

And unfortunately, this is one reason why Wikipedia hasn't really exploited its technical potential yet--it's being directed by non-tech people. People like Brion Vibber, who know MediaWiki in-and-out, should be the ones making the technical decisions--not Sue or Erik, to say the least. Sue is nothing more than a ruthless businesswoman--but when she dictates the WMF's tech directions, it usually ends up as nothing very special. The CTO could be responsible for so much more, given that the WMF has relatively few employees anyway but they have an enormous collection of tech and server stuff.

A very good point, indeed. The technology that drives the accumulation of publicly-contributed content on the largest and best-known crowdsourced site shouldn't be under the control of people who define success primarily in terms of how much "churn" they can generate, and the extent to which they can convince a few rich suckers that they're a "movement." Preferably, it should be in the hands of scholars and educators, or barring that possibility, actual technologists. The WMF is mostly a recruiting and fundraising organization (which just happens to call itself a "charity").

I suppose the counter-argument might be that too much software innovation might actually get in the way of contributors and contributions, but it's not like they've made much of an effort in terms of ergonomics and ease-of-use, either.

[Milton Roe]

A very good point, indeed. The technology that drives the accumulation of publicly-contributed content on the largest and best-known crowdsourced site shouldn't be under the control of people who define success primarily in terms of how much "churn" they can generate, and the extent to which they can convince a few rich suckers that they're a "movement." Preferably, it should be in the hands of scholars and educators, or barring that possibility, actual technologists. The WMF is mostly a recruiting and fundraising organization (which just happens to call itself a "charity").

I suppose the counter-argument might be that too much software innovation might actually get in the way of contributors and contributions, but it's not like they've made much of an effort in terms of ergonomics and ease-of-use, either.

There are good reasons not to have the ultimate-software-engineers in charge, since they are (after all) engineers who think like engineers. They would rather have the system be logical than easy-to-use, and that is why many engineers should be sentenced to use the technology of OTHER engineers, without benefit of even a poorly written user manual.

Consider the main claim-to-fame of Steve Jobs: which is that he's able to act in some cases like a one-man focus-group and beta-tester. You know, you could hire people for that....

Engineers are the sort who provide a gizmo that locks all the control surfaces of an airplane, but no failsafe that doesn't allow the thing to take off with them still locked. Engineers design a warning light and chime for every possible thing that can go wrong, with no heirarchy, so that in a bad situation in a iced-up jetliner (the recent Air France Airbus disaster, per its black-boxes just recovered from 12,000 feet down in the Atlantic), you can't tell the "low airspeed" and "stall" chimes from any of severa dozen other chimes, warning signals, and whatnot. And they do the same with nuclear powerplants (see the history of Three Mile Island).

I was reminded of how bad engineering for non-engineers can be the other day when I bought a (very clean and perfectly maintained) red 1998 Volvo as a backup car. Boxy, but good. Since it was for the US market, it had an 18 gallon fuel tank guage. MARKED IN GALLONS. Which is good, but not something you saw much in the first century of automobile making. When nearly empty, I put 12 gallons in it and it read 12.5 gallons. I've lived all my life with cars that (mostly) have not been able to tell you other than with the manual how much fuel your tank holds, or how much of it you have left (except as a fraction). Anything more than that, is a 21st century innovation.

If a software engineer had designed my Volvo in 2011, it would have one gigantic LCD touch screen on the dash, and no default readouts at all. After touching START it would give a tree with a dropdown menu, one of which would be "consumable fluids status." Happy day: "gasoline" is in alpabetical order before "oil" and "wiper fluid." Let me see, do I want "gallons" or "liters"? Press "unit set mem."

[EricBarbour]

If a software engineer had designed my Volvo in 2011, it would have one gigantic LCD touch screen on the dash, and no default readouts at all. After touching START it would give a tree with a dropdown menu, one of which would be "consumable fluids status." Happy day: "gasoline" is in alpabetical order before "oil" and "wiper fluid." Let me see, do I want "gallons" or "liters"? Press "unit set mem."

And very likely, you would have mysterious software bugs and crashes and data losses. Because it would be running a real-time OS like a "minimal" Linux distro or Wind River, all of which make the earlier mainframe OSes like Unix or OS/370 seem laughably primitive and tiny. Modern OSes run massive numbers of background services, especially if databases or other large datasets have to be maintained, used and updated.

This is why I gave up on programming in the 1980s--the happy, obscure, wasteful world of object oriented programming was upon us, and I was constantly being told that piling up layers of abstraction between the processor hardware and the code developer was a "good thing".

When I owned an Amiga in 1990, it ran a multithreaded operating system that was even better than what the Mac had at the time--and it did it in 1 megabyte of memory, with one floppy drive. Complete with GUI, color graphics, etc.
My shiny new Mac Mini has 2 gigabytes of memory and a 320 GB hard drive, and running more than three large applications on it at once is a nightmare. It barely handles OpenOffice. (I just checked the activity monitor---other than OS kernel and services, the only process running is Firefox, and it's consuming almost 450 megabytes of RAM.......)

And let's not forget what Zoloft pointed out 2 months ago: the official Wikimedia "disaster recovery plan".

This post has been edited by EricBarbour: Thu 9th June 2011, 9:21am

[EricBarbour]

And btw, the Mozilla developers know that Firefox wastes RAM.
Still waiting for them to do something.

[thekohser]

Erik Moeller (yes, that Erik Moeller) makes it sound like Danese Cooper was so easily replaceable, that she doesn't even need replacement!

"I’m taking on the role of VP of Engineering and Product Development,
on an interim basis for now. We’re not going to immediately hire
either for that role or a CTO role. Thanks to Mark, Tim and Brion, we
have very strong architectural leadership in the department. Moreover,
we’ve got more than enough disruptive change as an engineering
organization to absorb for now, so we’ve decided that it doesn’t make
sense to immediately bring in a new person to lead the department."

Enjoy your last few days of superfluousness, Danese!

...about why Brion Vibber would be willing to return, given that he'll be reporting to Danese Cooper and not Erik Moeller...

Guess who Brion's reporting to now?

This post has been edited by thekohser: Tue 21st June 2011, 5:34pm