Preston L. Bannister { random memes }

As an American with German and British ancestors, the following makes an odd sort of sense .

Stolen from: Incogblogo.net – comments on cheer bear

TRADITIONAL CORPORATION
You have two cows.
You sell one and buy a bull. Your herd multiplies, and the economy grows. You sell them and retire on the income.

AMERICAN CORPORATION
You have two cows.
You sell three of them to your publicly listed company, using letters of credit opened by your brother-in-law at the bank, then execute a debt/equity swap with an associated general offer so that you get all four cows back, with a tax exemption for five cows. The milk rights of the six cows are transferred via an intermediary to a Cayman Island company secretly owned by the majority shareholder who sells the rights to all seven cows back to your listed company. The annual report says the company owns eight cows, with an option on one more. Sell one cow to buy a new president of the United States, leaving you with nine cows. No balance sheet provided with the release. The public buys your bull.

AN AUSTRALIAN CORPORATION
You have two cows.
You sell one, and force the other to produce the milk of four cows. You are surprised when the cow drops dead.

FRENCH CORPORATION
You have two cows.
You go on strike because you want three cows.

JAPANESE CORPORATION
You have two cows.
You redesign them so they are one-tenth the size of an ordinary cow and produce twenty time the milk. You then create clever cow cartoon images called Cowikimon and market them worldwide.

A GERMAN CORPORATION
You have two cows.
You reengineer them so they live for 100 years, eat once a month, and milk themselves.

AN ITALIAN CORPORATION
You have two cows, but you don’t know where they are… You break for lunch.

A RUSSIAN CORPORATION
You have two cows.
You count them and learn you have five cows.
You count them again and learn you have 42 cows.
You count them again and learn you have 12 cows.
You stop counting cows and open another bottle of vodka.

A SWISS CORPORATION
You have 5000 cows, none of which belong to you. You charge others for storing them.

CHINESE CORPORATION
You have two cows.
You have 300 people milking them.
You claim full employment, high bovine productivity, and arrest the newsman who reported the numbers.

A BRITISH CORPORATION
You have two cows… both are mad.

A NEW ZEALAND CORPORATION
You have two cows… and the one on the left is kinda cute…

File these under “why you should not use XHTML” – why DOCTYPES not relevant? and Sending XHTML as text/html Considered Harmful.

My conclusion – always declare the DOCTYPE from HTML 4.0 for web pages, and ignore XHTML.

Frankly using XHTML is like filing all your bills alphabetically – it sounds like a good idea, but turns out not to be worth the time invested. After digesting a series of books on HTML 4, CSS, Javascript and DOM – I just don’t see any advantage to XHTML. Sure it is “nice” that as well-formed XML it passes validation (or at least can in theory). Since web pages are in almost every case served to web browsers, and browsers have excellent HTML parsers, there is no gain there. If you want to serve XML over the web as an API to non-browser clients, there just isn’t any reason I can see to use XHTML. Sure, XHTML is a new-and-improved cleaner version of HTML. This is “nice” but not important. Sure it would be “nice” if all the HTML on the web was XHTML (as this makes writing parsers easier), but this will never happen. As long as the majority of the web is HTML you will need HTML parsers. As long as you must have HTML parsers, the XHTML parsers are simply redundant.

There is a lot of function in HTML 4, CSS, DOM and Javascript. We have only recently been able to count on near-universal HTML 4 support, and the majority of the web does not take advantage of what is already there. I think it fair to say that as a profession we are still learning to take advantage of current available function. I am a long ways from convinced that XHTML or XForms or CSS3 (or the like) is really enough of a gain to be worth the trouble.

An interesting, and sobering story about Patents and Linux from Tim Bray.

After reading Gosling’s posts on Java and whether open-source makes sense when applied to the Java platform, I have to admit that my interest in open-source Java is a means not an end. I do think Sun’s control over Java in the present form is a problem, but I am not at all clear on the appropriate solution. Got stopped for a while on this point.

A bit later I ran across a discussion about Linux desktops and how Gnome was implemented in C, and KDE was using C++ code. Ack. My personal belief is that code of this sort should be written in something higher level (like Java) rather than C/C++, so this just strikes me as wrong. On the other hand the most likely alternatives are Java or Mono. Sun’s Java does not ship with most (all?) Linux distributions, and certainly is not a guaranteed part of the basic installation. Presumably this has something to do with Sun’s licensing. Mono is a great idea, but copying Microsoft seems a risky strategy.

Test #1:

Java is most useful when it is everywhere. One good end-to-end test for Java licensing would be if the Sun JRE shipped with all the mainstream Linux distributions as a part of the basic installation. Sun will need to adopt a whatever-it-takes attitude to make this happen.

Test #2:

Mono came about in part because Microsoft’s notion of a Common Language Runtime is indeed a step upwards. Sun’s JVM is good at supporting Java-like languages, and perhaps not much else. The ability to have near seamless integration between languages is, from a programmer’s perspective, extremely cool.

A good end-to-end test would be to host Perl on the Sun JVM, and do whatever-it-takes to make Perl on the (J)VM outperform the current Perl 5 interpreter. This entails some careful profiling and likely some small changes to the VM, and should obviate the need for “Parrot”. I fully expect the fancy JIT compiler in the JVM to out-perform both the Perl 5 interpreter and “Parrot”. Perl is sufficiently different from Java so that once the VM can support Perl well, likely all the mainstream languages will perform well.

Maybe we do not need open-source Java

Open-source Java is a means to make Java more prevalent, but what we are really interested in is the end result. The above two tests measure the prevalence of Java (or more exactly the JVM) in two independent dimensions. In the end it does not much matter which path Sun takes, as long as we get where we need to go.

Tim Bray asks an interesting question in The Back Door.

I remember a time where a Sun workstation was possibly the coolest thing to have on a programmer’s desk. Much of the most interesting new software and research was done on Sun boxes. Techies loved Sun and pulled Sun hardware in the back door whenever possible. Over time the Sun reputation percolated up to the front office, just in time for the 1990’s boom and .com bubble. Somewhere in there Sun became a well-respected vendor coming in the front door, and did terrific business.

Sun was once cool because if you wanted cutting edge workstation techology (bitmapped graphic displays, fast CPUs) there just wasn’t any real competition. BSD Unix had become nearly a standard in the academic and research communities, and Sun had a superbe version of BSD Unix. The fact that Sun eventually made pretty decent servers allowed the techies to stay with the company they loved, and push Sun hardware to the front office.

But as Tim’s note brings to mind – Sun is no longer cool. At one point in time I would have been thrilled to have a Sun box on my desk. Now – it simply seems like too much of a bother.

There are two obvious points where Sun is off-target. First, the Solaris version of Unix has become an off-brand, as for all practical purposes Linux has become the standard version of Unix. I know this statement freaks out long-time Solaris fans, but walk in the “back door” and you will find MS Windows and Linux. Pretty much everything else is foisted on the techies by the front office or out-of-touch old fogies. Who wants to learn a non-standard version of Unix?

( My teenage son would if given the opportunity point out the that I am one of the old guys, but that is just his opinion ).

The second point is Sun hardware at the desktop – is just not interesting. Generic x86 and Apple PCs are entirely sufficient, and a lot less hassle.

What would make Sun cool once again? Well, I would say that the Sun folks got handed an asteriod-size hint at JavaOne. From all reports Apple laptops outnumbered pretty much everything else! Apple laptops run Unix – a BSD Unix descendent oddly enough, and admittedly a “non-standard” OS to all those folks accustomed to Windows and Linux. Apple more than made up for the difference by providing a superbe complete package.

Warning – a herd of scared cows is headed for slaughter in the following….

For years (decades?) an Apple-Sun alliance has almost seemed to make sense. Both companies produce really very nice hardware. The bottom of Sun’s range pretty much marked the top of Apple’s range. They are even both located in the same general area. We may have finally reached the time where the “almost” part goes away.

The general notion is for Sun to lock in a clean upscale path from the coolest desktop around.

Now this presents some internal stumbling blocks for the Sun folks. The Sparc CPU will have to fade away. We don’t need Sparc on the desktop, and eventually only legacy customers will be interested in Sparc on the server. This means that Sun adopts use of IBM’s PowerPC CPU, starting from the desktop and eventually moving up to the server. I know that just this alone will cause a great deal in internal, er “discussion” within Sun.

Oooo … interesting side-thought. Given the current technology landscape Sun could gracefully bow out of the single-CPU performance race with Sparc. If you are not (on the server) especially interested in single CPU performance, that leaves the Sparc folk with just shrinking the chip (lowering cost and power use) as new fab technologies come along. Not much need for new architecture. Sun could well support their legacy Sparc customers with a much smaller R&D outlay.

There is a wildcard in the JDS desktop and Opteron server combo. Offering a good upscale server path to Linux crowd is a good bet. As to the Java Desktop – I just don’t know. Sun is historically good at coming up with some really excellent ideas, but somehow when it comes to desktop software fails to put out really polished implementations. Don’t know why this is – probably something in the company mindset.

As to open-source Solaris – I really want to care about this. Probably this has something to do with the fact at one point I would have killed for a Sparcstation 1 (figuratively speaking). I’ve even downloaded Solaris x86 ISOs, burned CDs … but just can’t quite see reason to spend time with Solaris.

Setup a Linksys 802.11b router and notebook and a bit over a year ago (a bit before 802.11g was quite standard). Worked pretty well – could use the notebook anywhere in the house, and both in the front and back yard.

Over the last several months the range and reliability of the wireless network dropped off badly. I suspect both my neighbors bought 802.11g hubs, as there were two new SSIDs visible when reception was poor. Some of the online reviews suggest that older 802.11b hardware may not co-exist well with nearby active 802.11g gear.

Add to this the fact that the Linksys router would die about once a day (both local and wireless) and would only come back up if power cycled. More than a little annoying.

Taking this as a hint, ordered a D-Link DI-624 802.11g router (based on good reviews), and an 802.11a/b/g mini PCI card for the notebook. The D-Link works well, and has some very useful features not offered by the old Linksys. The D-Link router came first, and once setup the network stayed up without incident. So far so good…

The mini PCI network card for the notebook came last. Opened up the notebook to replace the card, and noticed that one of the two antenna connections to the card was loose.

Oh.

Seems the limited range on 802.11b might just have been due to the loose connection. Had not thought to open up the notebook and look for loose connections.

Oh well. With the new card the range is better, reception is again reliable, and speed (which really wasn’t an issue) is indeed somewhat better. Plus the notebook will now be usable on any 802.11 a, b or g network, which might someday be of use.

Interesting take on IBM and Linux in IBM is in a Pickle (Again). Given this is a significant Sun guy talking, this might even be mistaken as Sun’s point of view (don’t take that last literally).

I do think Jonathon has a point – but I suspect he is a little off target.

On a semi-related subject, I have been playing with different Linux distributions again. Had to – the Redhat 7.3 box died I was using for the SMB domain controlller (with roaming profiles) and file server.

• Installed Fedora core 2 (the latest) on the box meant to be the new shared server. This was a mixed bag – some things went very well, and some things were extremely frustrating. The Redhat startup scripts are a continuing royal pain to deal with, as even when I know exactly what I need to do, puzzling out the tangled mess of scripts run at startup can take days before I can find where to plug in my changes. Shell scripts are a useful tool, but the performance freak in my nature looks at the endless startup scripts written by Redhat and wants to run away screaming. Running up2date chewed up pretty much an entire day, as I had to limit the updates to a few items at a time to prevent up2date from hanging.
• Installed Gentoo Linux on the fastest box in the house (I have delusions of doing my Windows development in a VMware session on a Linux box). There is something quite satisfying in compiling for my hardware with exactly the options what I want want. On the other hand I have not finished – X and Gnome are compiled but not yet configured. I really like the notion of compiling exactly for my hardware and configuration – and the “emerge” feature in Gentoo is quite impressive – though my “free” time is a little too finite.
• Installed Xandros Linux on a slower box (a mere 800Mhz CPU) after reading a recommendation on the web. The install was slick and went well. There was trouble once the desktop came up as for some reason the system would go completely non-responsive every few seconds.
• Installed Debian in place of Xandros. The Debian “netinst” install is not as flashy (no GUI) but is running well (nice to have a KVM switch at times like this ).

How does this relate to IBM, Sun, and Linux? Simple – there are too many Linux distributions, and it is easy to create a new Linux distribution. IBM can use the Suse distribution, or they can create their own. IBM could produce their own distribution, or could adopt a family of distributions. Do you need a near-stock installation for a small number of machines? If so IBM could put in and maintain Redhat, Fedora, Suse, or Debian. Do you want to squeeze every once of performance out of a large homogeneous collection of servers? If so IBM could use Gentoo to build a customized distribution exactly for your hardware and needs.

What is the point of all this? Redhat is not Microsoft. Linux cannot be made into a proprietary product. IBM has choices with Linux that were not possible with Microsoft. Customers don’t generally care which exact Linux distribution is installed on their machines. They do care about service. They do care if their applications run and run well. Otherwise if IBM promises support at a reasonable price, they should be perfectly happy with whatever IBM chooses to install.

Jonathon has a point – but it may not be the one he intended.

Hired a gardener to cut the grass a bit over a year ago. Have to admit that since I have not paid a lot of attention to the state of the backyard. A few weeks ago I noticied the north side of the house was getting soggy*, and there seemed to be a dry spot yard on the east side. Not big deal – I thought – the weather has changed and the sprinklers need adjusting. A couple weeks after changing the watering times on the sprinklers things were not improved, so I fired up the sprinklers, walked around, and found problems. Turns out I have at least two broken sprinkler heads, and at least one tree that needs to be removed. Oh fun.

A soggy yard means muddy dog footprints in the house, so I cannot ignore the problem…

This is Southern California. We get a fair amount of humidity from the ocean, but very little rain. Without that added humidity (and the imported Colorado river water) this area would be pretty close to hard desert. Lose the sprinklers and that part of your yard dies.

The house was built a bit over ten years ago and is located up in the foothills with the development bordered by a wilderness area. To form the lots they had to do a lot of the usual sculpting with an earthmover. The end result is that the ground is 1-2 feet of poor soil underlaid by very hard clay. Water cannot drain through the clay. On the plus side the entire community is very well drained, and even the heaviest rainfall disappears without trouble once it hits the street. Draining your own yard is your responsibility.

If you are putting in drainage into your yard, the usual practice of putting in visible surface drains makes remarkably little sense. When you to stop to think a bit you realise that for surface drains to work all the ground must be saturated up to the level of the drains(!). This makes no sense unless you are in an area where water soaks easily into the ground. My last house was like this – when you dug downward you found deep loose sandy soil. With that sort of soil you could dump seemingly any amount of water on the yard, and never get standing water (surface drains where not really needed). With ground like my present house proper drainage is critical. Surface drains are pretty near useless.

* The soggy side of the yard is in part due to my respect for the landscaper’s expertise – or as it turns out lack of expertise. I let the guy talk me out of putting in subsurface drains in the north part of the yard. That was a big mistake.

In contrast I had earlier put in subsurface drains on the south side of the house. The south side is slightly downslope, and just with normal watering the drainage from both my and my neighbor’s yard made that area into smelly muck. Since putting in subsurface drains that part of the yard is never a problem, and even with the heaviest rain that part of the yard never has standing water.

The back (east) side of the yard is at the bottom of a slope maintained by the community. I was concerned that drainage from the slope might be a problem. When the landscaper was working on the yard, I insisted on subsurface drains all along that edge of the yard. The drains in combination with a low wall on the downslope have worked very well. On the years with the heaviest rain (every few years we get about a month of relatively heavy rain) when my neighbors had problem with standing water, I did not.

In truth I was not satisfied with the sprinkler system as put in landscaper originally. He tried to divide the yard into too few zones, with the end result that one part of the yard would be near-dead for lack of water, and a nearby area would be muck.

Looks like I need to find another someone to work on the backyard – at least to fix the broken sprinklers, remove the offending tree, probably to put in proper subsurface drains in the north side yard, and possibly to re-do the entire sprinkler system.

Don’t know how to pick someone reasonable to do the work. There are leaflets dropped at the front door pretty much every week, so I guess that has to be the starting point.