LevelsOfDetail

I've been battling with the SFU NFS server for a while now. One mystery I haven't solved with it it's mapping of the 'group' owner of a file.

In Unix, every file has an associated user and group, and these are stored in the file system. In Windows, each file has an owner SID, which can either be a user or a group (at least, that is all that is revealed in the UI).

So here's the mystery: when the nfs client asks for the user and group of a file, how does the SFU NFS server decide? I found that if I set the owner of a file (on the windows side) to a user's SID, then the "user" part of the answer is the UID as specified in the User Mapping component. However, the group part always returns -2 which is convention to mean the nobody user.

Interestingly, if I create a file from the nfs client (Linux) with some UID and GID, the server accepts this information, and somehow stores it. Further accesses to the that file return the correct UID and GID, even across mounts and unmounts.

More annoyingly, a file created on the windows side by a user in the Administrator group always gets its owner set to the Administrators group SID. So when NFS servers up this file, it tries to map the Administrators group as a user, and always fails.

That aside, though, it is clear from the client-side file creation example that Windows is capable of storing the 'group' associated with a file somewhere. It's just that this information seems to be inaccessible and unmodifiable.

Anyways, getting this straight just became such a pain, that I decided to install Ubuntu on the machine instead.

Guerilla Youtube marketing I suppose.

• Google Browser Sync: A Firefox extension to sync browser settings. I primarily use it to sync my bookmarks across machines, and it works remarkably well. Only supports Firefox though.
• Japanese Fonts: A page with a bunch of links to freely downloadable Japanese fonts, screenshots included. The availability of high quality Japanese fonts has greatly improved over the past few years. With a full setup, you can even rival OSX rendering quality in a Linux environment.

I'm a fan of Proggy Programming Fonts for terminal use. But they don't work out of the box on my Ubuntu machine.

The reason seems to have something to do with my display's DPI setting, in combination with the fact that the TTF versions of these fonts are not real embedded bitmap fonts.

Actually, let me backup. The website linked above actually advertises three versions of each fonts: the windows bitmap font (.fon), the truetype file (.ttf), and a bitmap fonts for X (.pcf). Now normally, gtk+freetype+fontconfig is supposed to be able to pick up the .pcf file out of your ~/.fonts directory, but for some reason this doesn't work in Edgy.

Because of this, I resorted to trying to use the ttf version, but every gnome app I tried insisted on scaling the result. Because this font is unhinted, it looks terrible at any resolution other than the one it was intended for. Usually these types of fonts are done as a true type font with embedded bitmaps, but for some reason, the authors of this font decided to create outlines that approximated a bitmap font. If you open up the font in a fontforge and then view a scaled version, you'll see that each glyph is a series of vectorized squares that match the original pixels.

I can't imagine why they did this, but because of this, now my apps think its a regular scalable font. Worse yet, because my display DPI doesn't match some multiple of what is encoded in the font, it's impossible for me to specify the exact size (in points) that will make this font look right.

It was time to remind myself how to use fontconfig. I spent a lot of time realizing that the location of the user-specific font configuration file is ~/.fonts.conf and not ~/.fonts/.fonts.conf. But anyways, once that was figured out, the required bit is actually pretty small:

<fontconfig>
  <match target="font">
    <test name="family">
      <string>ProggyCleanTTSZ</string>
    </test>
    <edit name="pixelsize">
       <double>16</double>
    </edit>
    <edit name="antialias">
       <bool>false</bool>
    </edit>
    <edit name="hinting">
       <bool>false</bool>
    </edit>
  </match>
</fontconfig>

In layman speak, that's roughly: "If ProggyCleanTTSZ is returned as a query for a font, make sure its rendered with pixel size 16, and no antialiasing and no hinting." This forces every app that uses fontconfig to always render this font at pixel size 16 (which I determined was the intended size, after a little experimenting)

As is the case where I'm put in any situation where there are multiple options, I've switched blog software many times. Mainly between Wordpress, Textpattern, and now Moveable type.

While the migrations for most of these are fairly well supported (well, WP to MT is a bit of work), the one thing that gets you every time is non-text content. While the various conversion routines understand mostly how to convert database schemas and textual content, they fall over on other things such as images. This is mainly because there's no standard way to store images in databases, and not even all blog software have built in management of images.

If you want to engineer your blog such that you're not locked into a platform, I'd recommend you use an external image management system, and just use <img> tags in your posts. Those will always work.

One of the reasons I built a PC this time around was the price. The machine I built cost $1300 or so. The Mac Pro I could have bought would have cost at least $2000 (though with better specs).

But if you have the similar specs, the price ends up being pretty similar. Sure, I don't need fully buffered ECC RAM, or even quad core.

But I have this nagging thought: Is there some value in the fact that if I were to buy a Mac Pro, I'd stop worrying about what I could buy otherwise?

It's clear to me that my previous Powerbook was an incorrect purchase. My usage pattern doesn't fit the "big laptop" use case. I'm more of a "big desktop and small laptop" kind of person. And the PB was just too behind on the performance curve. But if someone handed me a Mac Pro, I'm fairly certain I'd be happy with it. Especially with the advent of virtualization on the Mac, I won't have any of the software annoyances that I had before.

I look at my recent HHK purchases. Though these things cost over $200 a piece, I've all but stopped looking at keyboards since I bought them. They were expensive, but given the amount of time I spent looking for the best inexpensive keyboard, could the time I "saved" from not having to continue searching have been worth the extra cost?

Does the guarantee of quality in product have value that's separate from the quality of the product itself?

Of course it's also crucial that you believe in the quality of the product. After using these HHK's for a month or so, I'm convinced that if you care about the feel of your keyboard, then they're certainly worth it. Likewise, the Mac Pro seems like a pretty good value for the price.

I got a sudden spike in my regular daily traffic last week, and it's all from Google, and all for an article that I wrote quite a while ago.

After googling a bit, I discovered that Google apparently does a full rebuild of their indexes approximately once a month. That means that even if it knows about your site, chances are if you're a small site, it won't pick up new pages for a while.

That means keeping your URL's constant is really important. If you're like me and you switch blog systems way too often, then you're destroying your articles even before Google can get to them.

Well it's been about four months since I sold the Powerbook. I had all types of reasons to get rid of it, but now I find myself looking at macs again.

What's changed? Well there are a few specific things:

• Mac performance got a lot better. The PPC to Intel jump was huge, and even Photoshop and friends are about to finish the transition to Intel code.
• VMware is going to release a product for the Mac. This makes most of the software incompatibility pain go away. I've been using a Windows desktop at home since I sold the mac, and I find that most of the time I need to run Windows apps is to interact with systems at work. But even then, it's just simple things like our VPN client (which probably works on the mac by now) and Outlook. All this stuff runs just fine in a VM.
• Vista was kind of a let down. A lot of stuff doesn't work with it. I'm not sure that they fixed any of the fundamental problems, and it kinda looks ugly IMHO. It feels like a mishmash of eye candy. It still has a lot of ugly greys, and a smattering of inconsistent designs. Sure, glass window borders are cool, but are they really useful? The real stuff I look forward to in Vista is real double buffering (which OS X has had for a while) as well as better font rendering.
• Messing with hardware is a pain. When I built my new desktop, I was super careful about the parts that I chose. I picked all the reputable manufacturers, and only picked pieces that were well reviewed. And yet.. I still had problems. The wireless on the Asus P5B is a little flaky, I get weird messed up graphics when waking up my machine sometimes.. resume from suspend take 15 seconds! I thought building at all would be worth my time again, but it's really hard to say that it was.
• The selection of keyboards is rather small for the Mac, but I settled on the HHK Pro, and that works fine with the mac

I have lots of time to think about it, since I'm not going to make any sudden moves. The other thing I realized that is somewhat unfortunate about PC's is that they're much harder to sell on craigslist. Apple's are recognizable, and have lasting value since they sport designs that appeal to everyone. How many people are you going to find that appreciate a custom built PC? Dell took over the market years ago. Nobody cares anymore.

I suppose all is not lost. I can still install Linux on my home machine and do something interesting with it. But I'm definitely feeling these days that I should have just spent a little more and gone for the Mac Pro.

I've come to believe that the number of the Service Pack for any Microsoft product actually is code for how they feel about the release.

• Initial Release = "Hey this doesn't crash for most people! Good luck getting any work done with it though!"
• SP1 = "Ok, all the really broken crap is fixed. But it's still now how it should be"
• SP2 = "This is about as good as it'll get. You can probably depend on it now"
• SP3 = "You really should upgrade to our latest product, but here's an update in case you don't want to pay us"
• SP4 = "Dude, this product is totally dead. Why are you even still using it?"

For different products, there might be a +/- factor of 1 SP release or so.

I'd point out that Apple loves to do the same thing. Someone should chart the MacOS 10.x.y releases in the same way. It seems like y is consistently greater than 5 before the series is retired.

The open source world is a little different. There, nothing ever reaches 1.0.

This post was inspired by the performance suckage that is Outlook 2007. I finally joined the downgrading bandwagon.