Isaac Rabinovitch: His Blog

I'd been meaning to get this blog going again, Before doing that, I wanted to finish moving the site to a new provider and work on the site as a whole. I just realized that's not going to happen before I move to Portland, and I really need to blog the move. I mean, jeez, it's a major change in my life. If I don't blog about it, why even have a blog?

So, why am I moving to Portland? Let's start with how I ended up in San Jose.

I grew up in Southern California, Fontana mostly. A long time ago I moved to Santa Cruz to go to UCSC. That didn't work out -- I have the dubious honor of flunking out of UCSC back in its granola days, when it had the reputation for being a hippie school nobody flunked out of.

Santa Cruz is sort of a suburb of Silicon Valley. Being a tech geek, I ended up with various high-tech jobs in both places. Seven years ago, I was dead broke and landed a job with Hurricane Electric, which has one of its data centers in downtown San Jose. (I could write an interesting blog post about why there are so many data centers downtown, and used to be a lot more. Another time.) I'd heard that downtown was a place where you could get by without a car, and I'd always wanted to give the pedestrian lifestyle a try.  So I rented a studio apartment in big Victorian about a mile from the data center.

Well, Hurricane Electric didn't work out, Since then, I've had jobs in Santa Clara, Menlo Park, and Oakland, If any of the these jobs had lasted longer, I might have overcome my intertia and moved to a bigger/close place. But I never did.
 
The pedestrian lifestyle thing turned out to be quite bogus. There are a number of good ammenities (my favorite being a pretty decent public/university library) and a lot of pleasant, tree-lined streets. But if you want to work, shop, or get entertained, you pretty much have to get in your car and drive somewhere.

The myth of a "pedestrian friendly" downtown comes from the huge effort San Jose expended to create one.  Starting in the 1980s, they built a transit mall to serve as a hub of the valley's brand new light rail system. They encouraged the development of  upscale housing, retail space, a big movie complex, grocery stores, fancy office buildings and other good stuff.

Didn't work out. The housing sold quickly enough (there's still a shortage of housing in Silicon Valley, even with the bursting of the dotcom and housing bubbles). But the folks who live in them have little to do with the day-to-day life of downtown -- when they need to shop or work they get in their cars and drive somewhere else.  The transit mall gets heavy use, but not by people with deep pockets, so the shopping never got close to economic critical mass. A lot of retailers closed shop, the movie theater closed (in the middle of the night, so the city couldn't stop United Artists from moving out its hardware). Retail space got converted to offices (at HE, I worked in what had been a dry cleaners), aggravating a glut of office space that's still pretty bad.

Despite the failure of the Pedestrian Utopia, downtown is not a bad place to live. Simple inertia might have kept me here a lot longer, except my for my landlords' desire for a little more privacy.

A lot of the houses on my street were built as boarding houses, catering to students going to San Jose State (which started out as a tiny teacher's college, with no dorms). Some of them got converted into apartments; my landlords own two of these, and live in part of the same one I live in. A little while back they decided to stop replacing their tenants as they moved out, so they could convert the extra space for their own use. There used to be three tenants, now there's just me. So they gave me a couple months notice to move out.

OK, fair enough, but it left me with a question to answer: where did I want to live now? See the next blog entry.

Related articles

• Should I base my start-up in Chicago, or move to Austin, Portland or Silicon Valley? (feldmanfile.blogspot.com)

One reason (possibly the reason) Microsoft produces so much bad software: the various parts of the company often work at cross purposes. As with any large organization, a lot of it is just poor coordination -- but former Microsoft VP Dick Brass, in this New York Times Op-Ed piece, claims that his own work on ClearType fonts and Tablet PCs was sabotaged
by people who put their own empire-building over the more obvious priority of producing good products.

I just got bit by this very syndrome after installing a lot of updates on my LE1700 tablet. (Irony alert!) One of these updates seems to have broken Windows 7's ability to identify USB devices and match them with the appropriate drivers. I need to roll the system back to an old restore point to verify this diagnosis. But System Restore won't let me do this until I repair some file system errors. So I schedule a boot-time repair (you can't repair the system drive while the OS is running). Which fails, because of one of these same updates. The repair utility suggests that I use System Restore to fix the problem!

OK, I know what to do next: boot a rescue OS and use it to repair the system drive. But not before I register my disdain with a blog entry...

Related articles by Zemanta

• Microsoft On Claims of Lameness: It's the Scale, Stupid [Microsoft] (gizmodo.com)
• An Early Account of How Microsoft Strangled Its Own Tablets [Microsoft] (gizmodo.com)
• Lost: If Microsofties Can't Live Together, Microsoft May Die Alone (seekingalpha.com)

The context of this post is Rob Weir's blog post about how Microsoft screwed up the implementation of www.browserchoice.eu.

That web site exists as a concession to EU anti-trust officials, who objected to the fact that Windows automatically uses Microsoft's own Internet Explorer web browser. This doesn't prevent users from installing and using other browsers in place of IE. But of course most users don't know they can do this, much less why they should. So, a European user who fires up Windows 7 for the first time is presented with the option of going to this web site and installing any of a dozen competing browsers. To insure fairness, the order in which the browsers are listed is randomized each time you access the site.

Except the randomness isn't as random as it should be. If you automatically bring up the site over and over (and of course many people have done that) it becomes obvious that certain placements are much more likely than others. In particular IE tends to come out at position 5 about half the time. Many folks believe this placement is somehow supposed to favor Microsoft.

The reality is another vindication of Hanlon's Law. It can be shown that this screwup reflects a common mistake in the way you write code that chooses a number randomly. Weir talks about the evidence for this and concludes that one should "reach for Knuth".

This is a reference to the Donald Knuth's well know text The Art of Computer Programming. Weir is arguing that whoever wrote this code needs to refresh their understanding of the science of generating random numbers.

Weir is himself making a mistake here. He assumes that all programmers are or should be computer scientists. The fact is most programmers don't have a thorough understanding of computer science, and don't need to. As with any technology, the technologists -- engineers, mechanics, etc. -- have to understand the scientist well enough to comprehend the structure of their work, but doesn't need to be able to reproduce the scientist's work. In fact, technologists who try usually end up being pretty bad scientists.

The Microsoft programmer's mistake was not writing a bad implementation of a random number algorithm. Their mistake was trying to write it at all. There's canned code for this sort of thing, and a good programmer relies on it.

Josh Bloch (who most certainly does rate as a computer scientist) refers to this very use case when he talks about why you should rely on library code.

Related articles by Zemanta

• Microsoft To Offer Users Choice Of Browser (mneylon.com)
• A Tale of Two Ballot Screens (opendotdotdot.blogspot.com)
• Microsoft's Ballot Screen is a Farce, Decoy (boycottnovell.com)

Just finished reading Kayla Williams military memoir, Love My Rifle More Than You: Young and Female in the U.S. Army. Only moderately informative or readable, but I have to share one excerpt.

Williams is in a Humvee in a convoy going from Tal Afar to Mosul. Convoys are prime insurgent targets, so  convoy duty is pretty nerve wracking. It doesn't help that Iraqis tend to be super-aggressive drivers -- and like bad drivers everywhere, they're blind to the risks they're running. Even if the risk is a nervous soldier with her M4 "on red" -- locked and loaded.

Now I'm noticing this local attempting to pass us. We're driving on the left, and this car is coming up fast on the grass beside us on the left.

It's gloomy and difficult to see, but it's clear that this car is planning to cut in ahead of us. I can tell by the way the driver is glancing back and forth that he is going to cut over.

This is making me nervous. It is also pissing me off. They should know we don't want them to do this. They should know better!

...

So I'm gesturing with my weapon at this car moving up next to us. I'm telling them: Do not cut us off! Back off! DON'T FUCK WITH ME!

The car is not pulling back. He is not getting the message.

It's tremendously loud with no doors. No one in the Humvee but me sees this car yet. It's my field of fire. It's my call to decide what to do next.

I raise my weapon and point directly into the car. I can feel my adrenaline pumping. I do not know what's going to happen.

I will shoot if this car is beginning to feel like a threat.

My weapon status is red. Always red on convoy now. My safety's on, though I know some soldiers are not bothering keeping their weapons on safety anymore. But I do. It's still less than a second. Flip, squeeze. After the first round, I can fire at will.

Just then a passenger on my side turns to look for the first time.

It's a little boy. Not more than eight or nine years old. I'm pointing my weapon at a boy who looks exactly like Rick's little brother.

The boy looks at me looking at him.

I lower my rifle and hold it with one hand across my knees. Without thinking, I wave at the boy with my other hand.

And after a moment he waves back.

Google Chrome is faster than any other browser. I mean, a lot faster. That's not a small thing when you spend as much time on the web as I do. (I distinctly remember being able to do technical writing before the web was invented, but I don't remember how.) And my main browser Firefox has real problems. 

Related articles by Zemanta

• Google Chrome Browser - Time To Make The Switch? (lockergnome.com)
• Top 10 Google Chrome Extensions for Productivity (taragana.com)

I started up picknit.com in 2003 on DreamHost. Like most budding web monkeys, I was long on plans and ideas, but short on time and motivation. It's only recently that I've started working on the site in earnest, but I always kept my personal email here. And email is really mission critical.

Which is why I ditched DreamHost in 2006. Overall, they're the slickest, most professional, and most cost-effective of the inexpensive hosting providers. But they keep having outages! There was a whole series of them in 2006; after several weeks of my email server being unavailable or sluggish as a beached whale, I moved on.

Alas, I've never found a hosting provider I've been really happy with. The low-end hosting marketplace is dominated by flakes and con artists. The serious professionals mostly don't find it worth their while to service anybody whose hosting budget is less than $100/month. My current provider, MediaTemple, is sort of an exception. For $20/month, I get shared hosting on a virtual server. If they have hardware downtime, they just move the virtual server to another system, and I see zero downtime.

But their support for customers at my tier is kind of half-hearted. It took me a while to figure out the configuration details from their sparse and inconsistent documentation. And their spam filter appears to be some homebrew nonsense that's not very tunable. That makes my spam white list crucial, and their control panel doesn't make white list maintenance easy.

So, I'm thinking of moving on. I thought of going back to DreamHost, but their outage problems remain unresolved. Liquid Web looks sort of promising, but that would mean going back to shared hosting on a physical server. They appear to be better at keeping their servers up than DreamHost, but still...

One solution is to host the site myself. My geek-friendly ISP, Speakeasy, gives me a free IP address, so it's doable without dynamic DNS gimmicks. I only have about 400 Mbps of upstream network capacity, but that will  do for now. If I ever start generating real traffic, I can rent some rack space somewhere.

But no, forget it. If my hardware fails my site is down until I straighten it out. Too much potential tsuris.

I've often thought of getting a virtual private server, which would end my dependence on the provider for getting the software features I need. I've balked at the cost (they start at $60/month) but perhaps I need to bite the bullet.

But wait! Here's an alternative: Slicehost is now offering VPSs for only $20 a month. The catch is that you're totally on your own for managing the thing. But maybe I want that. It fits in with the "learn by doing things the hard way" model that this web site is built around.


 

Related articles by Zemanta

• Webhost in five day server FAIL (go.theregister.com)

Started reading the Designer's Guide for MT4. As you might expect, Movable Type itself is used to maintain and host the content. People documenting Web CMSs always do this: use the product itself for the product docs. I suppose it's a natural thing to do, but it can be frustrating for the reader if the CMS isn't good at structuring content -- and Movable Type is not. There's only rudimentary breadcrumbing, and no automatic generation of "next" and "previous" links.

One of the challenges I've been facing as a tech writer is the growing preference for this kind of quick-and-dirty approach to documentation. I saw a lot of it at Sun, and the impression I got then was that it was part of the current infatuation with wikis and other crowdsourcing technologies. But I'm beginning it see that it also has a lot to do with the rise of Agile software engineering. If the software development cycle can be streamlined, deformalized, and drastically shortened, why can't you do the same with the documentation cycle?

Well, there are reasons, but I'm not going to get into it now. Suffice to say that I believe that documentation needs to be more structured, not less.  However, this is not an issue you can deal with through argument; the issues are just not understandable to most engineers. (Or a lot of tech writers, for that matter.) You have to demonstrate how structured methods make for better docs. And you have to do it in such a way that your formalized doc process fits in with the deformalized software development process.

It's a pretty problem. I suspect the answer might involve using both formal (XML, specialized authoring software, etc.) and informal content management tools. More on that later.

Zemanta has suggested a lot of related articles, based on what I just wrote. I'm going to insert links to some of the more interesting ones, without too much regard to their being ontopic.

Related articles by Zemanta

• Does Agile Development Matter? (devx.com)
• Agile Across the Enterprise: Prioritizing Value in Support and Training (tc.eserver.org)
• Blog Software Update (onemanandhisblog.com)
• Agile Software Development With Scrum (slideshare.net)
• Top Most Usable Open Source Content Management System- CMS Software for your organization (ceoworld.biz)
• TPM on Rails: Building a News App in Under Two Months (shhhaw.com)

I have little time for online language nazis. If somebody wants to verb a noun, what's the big deal? These rules are just conventions, more about establishing your place in the social structure than about communication. I use "good" language when formality  is required, and sometimes out of habit, but I try not to get religious about it.

One little problem: I myself have the compulsive personality of the worst kind of language nazi. What's worse, I know a bit more about language than most people, so I end up getting my buttons pushed by things other people never notice.

Case in point: hyphens and dashes. A long long time ago, I made my living typing academic papers, and had to learn all that nitpicky stuff about when to use hyphens and when to use dashes. Since typewriters didn't know how to print dashes, you used a double hyphen to represent it.

Nowadays, we all have fancy computer displays perfectly capable of showing dashes. Even so, people still use double hyphens. Mostly, this is because there's no dash key on most keyboards. But even newspapers and magazines, which use dash characters in their print editions, use double hyphens online. (This is a legacy of the early web, when there was no browser-independent way to represent non-ASCII characters.) It drives the typography nazi in me crazy.

And here's the last straw: if you manage to insert a dash character into a Movable Type post it converts it into a double hyphen!

Web sites exist primarily to present content and/or serve online software applications. On picknit.com, there's a third purpose: teaching myself web technology.

Which means I'll often do things the hard way. I could have started this blog on a free blog host (there are quite a few), but I wouldn't learn nearly as much about blogging software. So I've spent the last few days fiddling with various versions of  Movable Type blogging software, and finally settling on the Open Source version of MT 4.33. I'll explain that choice (and also describe my brief experience with WordPress) in a later post.

I'm pretty happy with this version of Movable Type, but I don't plan to stay locked into it. I probably won't bother with other software that's primarily about blogging (unless it has some feature I really want to check out) but I plan to fiddle with various Web CMS solutions, and most of those have blogging support that I'll want to try.

You have to sign in to comment. (Sorry about that, but I don't have time for the usual online snipers and flame warriors.) But you don't have to register. In fact you can't. Instead, you log in using your user ID from Google, Live Journal, and a few other portals and blogging providers. Or if you have an OpenID URL, you can use that. Just click the "sign in" link and follow the instructions.

To do list: learn enough about Movable Type to do some drastic customization of its look and feel. In particular, this fixed-width layout has to go.

BUG!!!! MT comes with a plugin application called Zemanta, which (among other things) helps you find graphics to go with your blog entry. That's how I found the photo above, which a certain Steve Rhodes has been good enough to release with a Creative Commons license. Zemanta also embeds an info link for the image, but somehow I managed to change that link to point to openid.net -- and Zemanta seems to have permanently re-associated that image with the new URL. Uncool! 

Now my Twitter followers (or rather follower) will be notified when I made a blog post.

(Later) Didn't work. Let's try it using my email address as a Twitter login.