Tuesday, January 24, 2012
CSS was a terrible mistake
There's always a flip-side. That's the madness of crowds.
That madness comes to mind as I contemplate the wreckage of Microsoft's SharePoint 2007 to Sharepoint 2010 Wiki conversion. How did this happen?
Part of the explanation is that Microsoft is a dying company, and SP 2010 was never finished. Microsoft redid their rich text editor, moving away from the IE embedded Word-like editor [1], but they didn't finish the conversion tools. Many formatted SP 2007 wiki pages cannot be edited in the new environment. (Apple does this stuff too, but with SP 2010 Microsoft has exceeded Apple's appallingly low standards.)
I think there's another factor though, and that's where the madness of crowds comes in. That was the decision made in the late 1990s to use Cascading Style Sheets (CSS) to control the layout of web content.
In retrospect CSS layout was a dreadful mistake, a madness of crowds "bridge-too-far" class mistake. Table based layout worked back when 100MB was a lot of memory. Mixing table models and CSS and style sheets is too much. Humans can't scale to this level of complexity; and browsers struggle mightily to reconcile multiple formatting and layout rules.
We are going to be paying for the CSS mistake for a very long time. We would be wise to at least recognize that we blundered, and try to understand why.
[1] Chrome and IE 9 are equally bad at editing our 540 or so partly-converted wiki pages.
Friday, January 13, 2012
CEO Compensation: Apple and more
Apple's Tom Cook isn't the worst example of excessive CEO compensation, but his huge stock option grant is bad news all the same.
It's bad news because it tells us that Apple 4.0 is becoming an average publicly traded corporation even more quickly than expected. This kind of compensation demoralizes employees who deal with tight budgets, even as their CEO earns a year of their pay in a day. Perhaps worse of all, when you pay a CEO a hundred times what their star employees receive, they imagine they're 100 times better than those employees.
An overpaid CEO is a CEO on chronic meth - intoxicated by the inarguable evidence of their perfect brilliance.
See also:
- Gordon's Notes: What are the consequences of extreme executive income? 11/2011
- Gordon's Notes: Natural selection and executive compensation 2/2009 - we've established an ecosystem that selects for internal predation, not wise leadership
- Gordon's Notes: Performance-based compensation and novel financial instruments: an explosive combination 1/2009
- Gordon's Notes: Will winner take all work? 9/2006
- Gordon's Notes: Mass disability and the middle class 9/2011.
- Gordon's Notes: Where has the money gone? To the very American oligarchy 2/2006
- Gordon's Notes: Unemployment and the new American economy - with some fixes 1/2001
Monday, January 09, 2012
Wisdom summarized - Elders speak
Elderly ‘Experts’ Share Life Advice in Cornell Project:
... Maintain social contacts. Avoid becoming isolated. When an invitation is issued, say yes. Take steps to stay engaged, and take advantage of opportunities to learn new thing...My takeaways:
- Some recommendations are in conflict. "Do what you love" as a career may well contradict "spend more time with your children".
- "chose to live each day as if it could be my last" is Jane Brody's personal one. Nobody else said that. Emily and I do something a bit different. We live as though our world were fragile and transient; both for us individually and for all we meet and know.
- Embrace aging, don't fight it. Yech. I suspect this is a form of denial. Physical/cognitive aging sucks. "Embracing" sounds like Stockholm Syndrome.
- Travel when young. Yeah, that's right. Good advice. Don't wait. Spend the bloody money; a fancy wheelchair isn't worth it.
Sunday, January 08, 2012
Do whipworms make dogs eat dirt?
The canine whipworm (Trichuris vulpis) is a common dog parasite.
The juvenile form of the parasite develops in dirt; it's transmitted by eating dirt (geophagy). It may be particularly common in dog parks [1]
Some dogs eat dirt.
Many parasites change the behaviors of their hosts to facilitate their life cycle.
So do whipworm infested dogs find dirt particularly tasty?
The association between geophagy and worm infection has been studied in humans, but a PubMed query on geophagy and trichuris vulpis in dogs produced no hits (PubMed includes veterinary literature).
I'm sure someone is studying this even as I write this post. I'm hoping for an article within the year.
[1] I don't know how hard it is to study this. A visit to the local dog park might be the basis for a great high school science experiment.
Rule 34 by Charlie Stross - my review
I read Charlie Stross's Rule 34. Here's my 5 star Amazon review (slightly modified as I thought of a few more things):
Rule 34 is brilliant work.
If Stross had written a novel placed in 2010, it would have been a top notch crime and suspense novel. Charlie's portrayal of the criminal mind, from silence-of-the-lambs psychopath (sociopath in UK speak, though that US/UK distinction is blurring) to every day petty crook, is top notch.
Stross puts us into the minds of his villains, heroes, and fools, using a curious 2nd person pronoun style that has a surprising significance. I loved how so many of his villains felt they were players, while others knew they were pawns. Only the most insightful know they're a cog in the machine.
A cog in a corporate machine that is. Whether cop or criminal or other, whether gay or straight, everyone is a component of a corporation. Not the megacorp of Gibson and Blade Runner, but the ubiquitous corporate meme that we also live in. The corporate meme has metastasized. It is invisible, it is everywhere, and it makes use of all material. Minds of all kinds, from Aspergerish to sociopath, for better and for worse, find a home in this ecosystem. The language of today's sycophantic guides to business is mainstream here.
Stross manages the suspense and twists of the thriller, and explores emerging sociology as he goes. The man has clearly done his homework on the entangled worlds of spam and netporn -- and I'm looking forward to the interviewers who ask him what that research was like. In other works Stross has written about the spamularity, and in Rule 34 he lays it out. He should give some credit to the spambots that constantly attack his personal blog.
Rule 34 stands on its own as a thriller/crime/character novel, but it doesn't take place in 2010. It takes place sometime in the 2020-2030s (at one point in the novel Stross gives us a date but I can't remember it exactly). A lot of the best science fiction features fully imagined worlds, and this world is complete. He's hit every current day extrapolation I've ever thought of, and many more besides. From the macroeconomics of middle Asia, to honey pots with honey pots, to amplified 00s style investment scams to home foundries to spamfested networked worlds to a carbon-priced economy to mass disability to cyberfraud of the vulnerable to ubiquitous surveillance to the bursting of the higher education bubble, to exploding jurisprudence creating universal crime … Phew. There's a lot more besides. I should have been making a list as I read.
Yes, Rule 34 is definitely a "hard" science fiction novel -- though it's easy to skip over the mind-bending parts if you're not a genre fan. You can't, however, completely avoid Stross's explorations of the nature of consciousness, and his take on the "Singularity" (aka rapture of the nerds). It's not giving away too much to say there's no rapture here. As to whether this is a Rainbow's End pre-Singular world … well, you'll have to read the novel and make your own decision. I'm not sure I'd take Stross's opinion on where this world of his is going - at least not at face value.
Oh, and if you squint a certain way, you can see a sort-of Batman in there too. I think that was deliberate; someone needs to ask Charlie about that.
Great stuff, and a Hugo contender for sure.
If you've read my blog you know I'm fond of extrapolating to the near future. Walking down my blog's tag list I see I'm keen on the nature and evolution of the Corporation, mind and consciousness, economics, today's history, emergence, carbon taxes, fraud and "the weak", the Great Recession (Lesser Depression), alternative minds (I live with 2 non-neurotypicals), corruption, politics, governance, the higher eduction and the education bubble, natural selection, identity, libertarianism (as a bad thing), memes, memory management, poverty (and mass disability), reputation management, schizophrenia and mental illness, security, technology, and the whitewater world. Not to mention the Singularity/Fermi Paradox (for me they're entangled -- I'm not a Happy Singularity sort of guy).
Well, Stross has, I dare to say, some of the same interests. Ok, so I'm not in much doubt of that. I read the guy religiously, and I'm sure I've reprocessed everything he's written. In Rule 34 he's hit all of these bases and more. Most impressively, if you're not looking for it, you could miss almost all of it. Stross weaves it in, just as he does a slow reveal of the nature of his characters, including the nature of the character you don't know about until the end.
Update: In one of those weird synchronicity things, Stross has his 2032 and 2092 predictions out this morning. Read 'em.
Saturday, January 07, 2012
Divorcing Google - and hoping for GoogleMinus
Google hired a hit man, and was shocked to find bodies. Now the remnants of Google 1.0 are punishing Google 2.0, though Matt Cutts hasn't said anything. He didn't used to be so quiet.
Google 1.0's glorious Data Liberation Front is pretty quiet too. Their Twitter feed went silent on 9/15/2011.
It's sad. I loved Google from my first searches in 1997 until the Google-Apple war began in 2009. Even then I wanted to believe in their original mission to free the world's knowledge. I didn't really lose faith until Google deleted my Reader Share memory - with 1 week's notice.
Yeah, I was denying simple arithmetic. Google's is an algorithmic corporation that iterates on its goals, and its goals are to delivery value to their paying customers. Advertisers. Yes, Google, like health insurance corporations, has an Agency problem.
Notice how much junk their is in Google searches these days? How many "plus.google.com" pages? How ads are growing across Google's search pages? This isn't going to stop -- not unless Google divides into two companies.
Now I'm moving off the Google platform. It's a slow and painful process. Just as painful as moving from DOS to Mac Classic, from Mac Classic to Windows, Windows to OS/2, OS/2 to Windows 2K, Windows to OS X, Palm to iOS…
Especially like moving from Palm to iOS. That's because there was a wasteland between the end of PalmOS and the rise of iOS. It was a kind of technological winter; a gap between one life and the other. I nursed my aging Palm devices along because the alternative was to resurrect my Franklin Planner.
That's what life is like post-Google - because there isn't a good alternative to Google. iCloud? Please. Microsoft? I wish. Yahoo!? Ok, you get the point. I won't go on.
So it's a tough divorce. It would be even harder if I were an Android customer. I wonder if Android users understand how tightly they're tied to Google -- and why.
How will this winter end? Amazon? Apple? Microsoft?!?
Or … perhaps …. GoogleMinus.
GoogleMinus, because there are businesses in Google that make money selling value to users. Google Docs, for example, sells ad-free solutions to schools and corporations. These aren't not huge businesses, but they might be profitable if they could lease infrastructure from GooglePlus.
GoogleMinus, because there are still, I think, some idealists left at Google -- and they might prefer to work for GoogleMinus rather than GooglePlus.
It could happen. The EU might require a breakup. Civil strife within Google might make a breakup internally acceptable.
Imagine a future when GoogleMinus packages GooglePlus search. For an annual fee of $100 I get the ability to block plus.google.com searches, and a control that lets me filter out web sites that run ads. Wouldn't that be interesting?
Update 1/9/12: This Android retrospective is relevant. I'd forgotten the day Google made its deal with Verizon; a deal signed in blood at midnight
Update 1/10/12: Today Google dedicated their search function to promoting Google+ properties.
Does earlier menarche explain the growing girl-boy academic gap?
Firstly, the age of menarche in Western nations has been declining...
Secular trends in age at menarche... [Paediatr Perinat Epidemiol. 2011] - PubMed - NCBI
Menarcheal age decreased over time in Western countries until cohorts born in the mid-20th century. It then stabilised, but limited data are available for recent cohorts. Menarche data were collected retrospectively by questionnaire in 2003-10 from 94,170 women who were participating in the Breakthrough Generations Study, aged 16-98 years, born 1908-93 and resident in the UK. Average menarcheal age declined from women born in 1908-19 (mean=13.5 years) to those born in 1945-49 (mean=12.6 years). It was then stable for several birth cohorts, but resumed its downward trend in recent cohorts (mean=12.3 years in 1990-93 cohort). Trends differed between socio-economic groups, but the recent decline was present in each group. In conclusion, menarcheal age appears to have decreased again in recent cohorts after a period of stabilization….
Surprisingly, we don't have good recent data - this is the best I could find. There's no evidence of a similar shift in males.
Secondly, in athletics, we know a maturational advantage of a few months can make a big difference in competitive events.
High school education is a competitive event. So, if girls are maturing earlier, how does this impact their academic performance relative to boys? Does this explain why our middle and senior high school honor role events are all female? Do boys get discouraged because they can't compete?
Seems like an interesting question. I'm thinking an 'all boy' school might be a good idea for our middle son.
Another one of our victims speaks
Another victim of America's Inquisition:
Notes From a Guantánamo Survivor - Murat Kurnaz - NYTimes.com
… was taken to Kandahar, in Afghanistan, where American interrogators asked me the same questions for several weeks: Where is Osama bin Laden? Was I with Al Qaeda? No, I told them, I was not with Al Qaeda. No, I had no idea where bin Laden was. I begged the interrogators to please call Germany and find out who I was. During their interrogations, they dunked my head under water and punched me in the stomach; they don’t call this waterboarding but it amounts to the same thing. I was sure I would drown.
At one point, I was chained to the ceiling of a building and hung by my hands for days. A doctor sometimes checked if I was O.K.; then I would be strung up again. The pain was unbearable...
I'm hoping one of these victims will get their case heard in the World Court.
There's hope, as always, in the younger generation. Maybe, one day, they'll hold trials (via Doc Searls …)
Otherwise occupied: What price revolution? | Hal Crowther | Independent Weekly
… what I think I see, through the media fog of polarized America, is the return of the full-fledged idealists (as opposed to single-issue idealists) who seemed to go underground around 1980, possibly because the mass media abandoned them during the mudslide of self-celebration that began with Reaganism and culminated in Facebook.
I say God bless them, and God will if he still has any investment in the United States of America. The Goliath they challenge has crushed a thousand Davids. The good news is that "the kids" are right on target. Their diagnosis is bull's-eye correct, and the patient is critical. For this country to survive, it must find saner ways to pursue and multiply wealth, and find them quickly. The cannibal capitalism that produced a Goldman Sachs and a Bernie Madoff is subhuman and obscene. There's no form of government more inherently offensive than plutocracy—only theocracy comes close. When a citizen comes of age in a plutocracy, he has no moral choice but to slay Pluto or die trying...
Friday, January 06, 2012
Can Apple produce first class software?
I use most of what Apple makes. Their hardware design is top notch. Their iOS and OS X systems are better than the competitions. Nobody beats them at retail or operations. Their service was always pretty good, but I think it's improving.
After that, things get dicey. Consider iWork ...
C’mon Apple, upgrade Numbers! | ZDNet
… Numbers performance is anything but speedy. In fact, it’s abysmal. I have a relatively tame no-formula five column table with 6000 rows and changing a value in a single cell requires 20 minutes to update the graph. That’s 20 minutes – not 20 seconds. With all the processing horsepower of a MBP, that’s absurd...
I'm amazed performance is that bad, but I've never done anything serious with Numbers.app. I have done a bit more with Pages.app, enough to suspect it doesn't scale well either. Overall iWork is an excellent start -- that needed a major follow-up in 2011.
Reading this, I thought about the other Apple products I use. The latest release of iPhotos is dismal. Aperture doesn't seem to try to compete for the pro market any more. Final Cut Pro is being abandoned by power users.
Their bundled apps are similarly miserable. ICal. iChat. Address Book. It's a veritable Hall of Software Shame. Safari is stabilizing again, but its performance issues have driven power users to Chrome on OS X.
The only Apple product that seems to rule its niche is iTunes.
Really, Apple is no better at applications than it is at Cloud services. If it didn't own the OS, none of its products would survive in the marketplace.
This is a Steve Jobs legacy. Tom Cook can't do worse. I'll close my eyes and cross my fingers and hope Cook's Apple can do better.
Tuesday, January 03, 2012
Great graphics: Apple OS and Video Codes
AppleInsider's the next ten years of Mac OS X has fairly predictable text, but the graphics are terrific. I particularly liked the Apple/NeXT/Unix graphic, but the video codec illustration is also fine.
I remembered A/UX, but not the Unix-based OS from the early 80s.
Monday, January 02, 2012
America's Inquisition
The Atlantic doesn't have this article online yet ...
Torturer's Apprentice, Cullen Murphy Jan/Feb 2012
... The first technique used by the Inquisition was ... torture by suspension ... it has been employed in the interrogation of prisoners in US custody ...
... The second technique was the rack ...
... The third technique involved water ... fabric plugged a victim's upturned mouth ... upon wich water was poured ...
... the (Bush) administration's threshold for when an act of torture begins was the point at which the inquisition stipulated that it must stop...
To clarify the last excerpt -- the Bush administration interrogation policy was remarkably similar to the Inquisition's interrogation policy. Their torture and our non-torture had similar (theoretical) stopping points.
ISP travails: Forget buffer bloat, what about my upload speeds?
Cringely was right about BufferBloat. I give him full credit for that one - I've not seen anyone else describing this networking problem. Once again I miss BYTE; they would have caught this long ago. (Old guy thing - I'm fascinated by good stuff that goes away without replacement.)
That's not what bugs me though. My video streaming is good enough. I'm a lot more concerned about my upload speeds. I'm getting about 0.1 mbps (real data) over CenturyLink DSL (formerly qwest) with a 1 GB cutoff [3], even though my download speeds are about 10mbps (networking traffic, not real bits). Yes, my current upload speeds are less than 5% of download speeds. CenturyLink has boosted its download speeds to be more Comcast competitive, at the hidden cost of upload speed.
No, I'm not running a Torrent. I've got too much to lose to be pirating movies, besides, it's wrong [1]. I'm just trying to share 1.5 GB of our summer vacation photos [2].
So what's the problem? I can just switch ISPs right? I'll just Google those sites that compare upload speeds. I remember using them only five ... ok ... ten years ago.
Not so fast. I can't find those handy comparison sites any more. The American ISP industry has consolidated so much there's not much point in doing comparisons. Locally my only alternative option is Comcast, and while it's probably better that CenturyLink it's not much better.
Cue the ominous music, because there are logical reasons for uploading to suffer in today's market. Most customers don't care about it, or don't realize that speeds are asymmetric so are baffled by mpbs marketing. On the other hand, Torrents are enemy #1 for most ISPs. Photo sharing and online backup services may simply be collateral damage of anti-BitTorrent wars.
I can live without sharing full sized images; almost nobody I share with wants a 6MB JPEG. I can also go without online backup -- I don't trust online backup anyway.
Even so, it's a bummer. Monopoly sucks.
[1] Sort of wrong. Given the legislative conduct of the IP industry there's a definite Robin Hood angle to stealing movies.
[2] Big JPEGs. A downside of today's monster DSLs, combined with some unfortunate side-effects of processing JPGs in Aperture.
[3] Throttling? Automated nightly modem reboots?
Update 1/7/12: I did hear back from CenturyLink. Basically, 300 kbps (100 kbpbs real throughput) is as good as it gets. If I want a different balance of upload and download then I need a "business" line. Sadly, this makes sense. I'm more like a business customer than a consumer customer.
Sunday, January 01, 2012
The GOP Primary is over now. What will the media cover next?
The 2011 GOP primary has to be one of the most foregone conclusions ever.
Once Perry flamed out, it was going to be Romney. Yes, there was great entertainment from Cain and Santorum and Gingrich and Bachman -- but really, they never had a hope. Maybe Gingrich for VP?
The numbers have fallen in line, and now it's all over but the formalities.
So what will the media cover next? I suppose it will be Romney's VP pick.
Has Microsoft lost the malware war?
I thought of John Halamka was a fairly careful writer, so this comment caught my eye (emphases mine):
Life as a Healthcare CIO: The Joy of Success
... One CIO received a negative audit report because new generations of viruses are no longer stopped by state of the art anti-virus software.... No one in the industry has solved the problem...
He refers to a previous post ...
Life as a Healthcare CIO: The Growing Malware Problem
... A new virus is released on the internet every 30 seconds. Modern viruses contain self modifying code. The "signature" approaches used in anti-virus software to rapidly identify known viruses, does not work with this new generation of malware.
Android attacks have increased 400% in the past year. Even the Apple App Store is not safe.
Apple OS X is not immune. Experts estimate that some recent viruses infections are 15% Mac...
Ok, so those sentences are a huge hit on his credibility. App Store issues are in no way comparable to Android attacks, and that 15% number could only be true for Microsoft Office malware (Duqu attacks a TrueType font parsing engine), or for something none of the Mac guys I read have run into. Nobody I know in the Mac community uses antivirals - even now. The cure is, for the moment, worse than the disease.
So Halamka is a bit lost, but it is true that the Stuxnet and Duqu platforms are formidable [1]. That's presumably what Halamka is talking about, and what some CIOs are thinking.
I haven't seen this elsewhere, but I don't track the Windows world all that closely. This will be something to watch over the next few months ...
[1] Even OS X Lion is no more secure than Windows 7 (for now). The only reason those viruses aren't attacking OS X machines is because there's no money in the Mac world. If Macs were used in banks they'd be at least as vulnerable to Duqu as Windows. The future (next?) version of OS X is expected to, like iOS, run signed code only.
Medical fads - are they cycling faster?
We've always had crazy fads in medicine.
I fell for a few when I had wet ears. Magnesium Sulfate post-MI is the one I remember best. That one even made it to textbooks before it died.
It's typical of medical fads that they infest journals, and now newspapers, but usually die before they get to textbooks. Estrogen for osteoporosis wasn't in that class -- that was a somewhat understandable research problem. Medical fads are less forgivable; they really aren't supported by evidence. They're built on easy money and bored specialists.
It feels like the fads are cycling faster. Emily and I thought the Vitamin D craze had another year or two, but it died fast.
Our local minor neurotrauma ("acute mild head injury") craze reeks of fadism. In Minnesota recommendations are being written into law, with little basis in science. As of today, PubMed has precious few studies.
Maybe it will be real. Some cults become established religions, some fads become science.
I don't think this one will make it to science, I do think it will cause significant harm along the way. Labels are powerful.
Hope this cycles as fast as Vitamin D, but putting minor traumatic brain injury into law may stretch its lifespan. Medical fadism is a crime against the vulnerable...
Update: More on the Vitamin D fad.
A few readers asked me for more detail on the vitamin D fad.
Briefly, for a year or two, I couldn't avoid popular articles claiming that Americans suffered from an epidemic of Vitamin D deficiency causing a wide range of disorders, and that recommended daily allowances were inadequate. Then, at the end of 2010, the Institute of Medicine published a report declaring that the science wasn't there, and that overdosing was more harmful than expected ... (emphases mine)
... The committee provided an exhaustive review of studies on potential health outcomes and found that the evidence supported a role for these nutrients in bone health but not in other health conditions. Overall, the committee concludes that the majority of Americans and Canadians are receiving adequate amounts of both calcium and vitamin D. Further, there is emerging evidence that too much of these nutrients may be harmful...
In retrospect, within a few months of the IOM report, the media attention ended. The fad moved on.
There's still science to be done of course. Ever since medical school I've wondered about the relationship of latitude to multiple sclerosis, and whether there was some kind of cutaneous immunity/solar radiation component. Today there are many interesting articles on the relationship between vitamin D and MS. That's research though, the fad is over.