The Itanium was announced in 1994 and so it is at least 8 years old. It is still a minor player in the larger processor chip market. The New York Times has a long article on the Itanium entitled Intel's Huge Bet Turns Iffy (free registration required). Gordon Bell (chief architect of the VAX architecture while at DEC) thinks Intel threw too many ideas into the design. Google doesn't want to use it because it consumes too much power:
Eric Schmidt, the computer scientist who is chief executive of Google, told a gathering of chip designers at Stanford last month that the computer world might now be headed in a new direction. In his vision of the future, small and inexpensive processors will act as Lego-style building blocks for a new class of vast data centers, which will increasingly displace the old-style mainframe and server computing of the 1980's and 90's.
It turns out, Dr. Schmidt told the audience, that what matters most to the computer designers at Google is not speed but power — low power, because data centers can consume as much electricity as a city.
If power efficiency does indeed trump processing speed, everything that Intel and Hewlett-Packard have done to pack raw power into the 221 million transistors of the new Itanium 2 could now be a handicap. The chip, which is as large as a silver dollar and whose 130 watts of power dissipation are enough to fry the proverbial egg, is not even a contender in the Google universe. "We're incredibly, incredibly power sensitive, and we've been talking to Intel about that," Dr. Schmidt said.
Wow, 130 Watts is a lot. AMD Athlons up around 60 to 70 watts are awfully hot and slightly hotter than equivalent throughput Intel P4 chips. Intel PIII offers a much nicer ratio of throughput/power and PPC is way lower in power. Anyone know what the power usage will be for AMD's forthcoming Opteron? Or how about for IBM's PPC chip that they are going to do for Apple?
I don't know where I've been but I only just found out that there are satellite radio services and that these services even work in cars. This strikes me as great for people who spend a lot of time on the road. It also strikes me as great for those in lousy reception areas, those who simply want more choices, and, since many of the channels have no commercials, those who hate commercials.
Among its stations XM has 3 classical stations with no commercials, 11 rock stations (7 commercial free) with different themes, 11 Hits stations (almost all with commercials), 6 that are focused on various previous decades, and 7 Jazz & Blues (6 commercial free). I find the only 3 classical stations to be disappointing. A blurb from their site:
One big idea can change everything. And XM Satellite Radio is one big idea: Radio to the Power of X. America's most popular satellite radio service gives you the power to choose what you want to hear - wherever and whenever you want it. XM offers 70 music channels - more than any other satellite radio service. Plus 30 channels of news, talk, sports and entertainment. 100 basic channels in all, for a low $9.99 monthly subscription. And now, XM is the first satellite radio service to offer a premium channel for an additional monthly fee.
Here's the XM Radio Full Channel Listing.
Sirius charges $12.95 per month. Sirius also has only 3 classical channels and mix of other channel types that is roughly similar to XM Radio. They claim to have 60 commercial free music channels and 40 other channels. Sirius appears to have Bluegrass similar to XM but also a Folk channel that XM doesn't have. XM has less talk, news, and sports. XM does have 3 comedy channels while Sirius has only 1.
The full Sirius channel listing is here.
For anyone who wants to know how much various portions of the P4 architecture contribute to the performance of an application this article has a link to a utility that, among other things, lets one turn second level cache prefetch on and off.
Between January and December 2001, companies hired 2.1 million IT workers, compared to 1.6 million workers between July 2001 and June 2002. Hiring dipped 25 percent during this tracking period.
Brains are competing across borders:
"The real challenge is offshore programming—not the few thousand [IT workers] that come to the U.S., but the workers in Ireland and South Africa and India that are paid much less to do the work," said Miller. "I think there is more work going offshore in part due to the pressure to keep costs down, and there's huge downward pressure on software vendors to keep their labor rates down," he added.
I don't mind ads that are images. Its the flashing that I can't stand. The flashing makes it hard to concentrate. What I really want is to be able to right click over an ad and be able to choose an option called "Stop Animation". Mozilla 1.1 doesn't have this option but it does have another: Right click over a flashing ad and choose the option "Block images from this server". Then reload the page. This just worked to make a flashing ad disappear for me the first time I tried it.
Of course some sites constantly change the ad URLs and have many different servers with different URLs serving ads. So this method is not perfect. But it looks like it can at least reduce the amount of flashing one has to experience.
You can download Mozilla v1.1 here.
I've been looking at visitor logs and am amazed to see MSIE 4.0 and 5.0 showing up. If you are running a fairly old IE version and want to upgrade to a newer version (and there are good security reasons for doing so) then an easy way to do that is to go to the Windows Update page on the MS site and go thru the fairly automated update process. It also will analyze your machine and tell you what security updates they have for it that you can have installed automatically. Its a wise thing to do. You shouldn't be out cruising the net with all sorts of old security holes unpatched.
Note as well that IE6 has sp1 out and IE5.5 sp2 has security patches that go beyond the sp2 level.
One other thing: I do not test my sites against anything older than IE5.5 and have no idea what IE 5.0 or 4.0 does with the style sheets that are used on the site. One person gave me a rather sketchy report of my blogs looking really bad on Mac IE 5.0. I have no way to test that and would suggest that anyone at IE5.0 or earlier upgrade (or install Opera or Moz).
I do most of my work on this site using Mozilla 1.1. I also check it against Opera 5 and IE 5.5. If you want to find Moz or Opera then check out my earlier link about Mozilla and other browsers.
There appears to be a design limitation for long running loops. If you have any designs that require continuous operation for, say, 50 billion years you are going to need to optimize and cut the time down. To be on the safe side your program is going to need to complete in less than 10 billion years:
But two new studies by Stanford University cosmologists suggest that it may be time to rethink this popular view of a "runaway universe." Instead of expanding exponentially, our cosmos may be in danger of collapsing in a "mere" 10 to 20 billion years, according to the Stanford team.
"The standard vision at the moment is that the universe is speeding up," said physics Professor Andrei Linde, "so we were surprised to find that a collapse could happen within such a short amount of time."
Linde and his wife, Renata Kallosh – also a professor of physics at Stanford – have authored two companion studies that raise the possibility of a cosmic "big crunch." Both papers are available on the physics research website, www.arXiv.org. "We tried our best to come up with a good theory that explains the acceleration of the universe, but ours is just a model," Linde noted. "It's just part of the answer."
If the Linde-Kallosh model is correct, then the universe, which appears to accelerating now, will begin to slow down and contract. "The universe may be doomed to collapse and disappear," Linde said. "Everything we see now, and at a much larger distance that we cannot see, will collapse into a point smaller than a proton. Locally, it will be the same as if you were inside a black hole. You will just discontinue your existence."
I hope this obvious design bug in the universe doesn't break too much existing software.
I was just over at Tim Blair's blog on Blogspot and saw him mentioning some Blogspot problem.
Here's what I do not understand: once a blogger becomes fairly well established and popular (and Tim Blair gets linked to by a lot of other bloggers) why stay with Blogspot? Blogspot seems to have enough troubles. I can see why its a good place for beginner bloggers and bloggers who get little traffic. But it only costs $11 a month to host on Hosting Matters for example. Now, one still has to go the work of setting of and configuring MovableType. But seems worth it to avoid all the Blogspot broken link problems and other problems that I keep seeing mentioned by and about Blogspot bloggers.
UPDATE: Here's Charles Johnson talking about Blogspot weirdnesses.
ANOTHER UPDATE: Excellent blogger Iain Murray can't get e-mail from 10 PM to 8 AM and weekends. Plus, he's on Blogspot. News flash for Iain: Some of the hosting services for real bloggers provide fairly reliable POP e-mail accounts as part of their basic plans.
Elliotte Rusty Harold has written an article 10 Reasons We Need Java 3.0 about why Java needs some big improvements. I agree it needs some big improvements. I even agree with a few of the ones he listed. I even have my own list of desired improvemnts that I'll list in future posts. But first lets go over some of his recommendations.
I'm fairly indifferent to his items 10 thru 4. Actually, that is not entirely true. I'm to varying degrees at least partially opposed to a few of them. One leaps out in particular: I'm opposed to deleting all the deprecated methods. Why? Leave aside that this breaks old code (though that's a good reason). Some of them are still necessary.
Necessary methods that are deprecated? Well, yes. When I compile my source code with JBuilder v5 (which is at JDK 1.3 level; no I haven't upgraded to JB v7 yet) it gives deprecated methods warnings. Here's an example: In java.sql.Date I use getYear(), getMonth(), and other similar methods. Well, calling them generates deprecation warnings. Why? I went and looked in them and they call getField(). But getField() is private. So I can't call it. So then I looked at the Javadoc comments for getYear and I see that it has been deprecated because we are supposed to use Calendar.get(Calendar.YEAR). But in order to do that we have to use Calendar (or really GregorianCalendar). But wait, I'm using java.sql.Date because I'm calling JDBC libraries that want their dates in java.sql.Date types. So why deprecate a class that Sun's own JDBC spec requires us to use?
I'm looking at the source code to the JDK 1.3's version of PreparedStatement. It doesn't have a simple straightforward setCalendar() method. It does have some sort of way of using one version of the setDate() method that you pass both a java.sql.Date and a Calendar object to. But from the poorly written description it sounds like the Calendar object may just have the timezone info to use and may not have the actual date value that you want to set in the database field.
In any case, if you have a java.sql.Resultset (ie if you are getting data back from databases using JDBC calls - but hey, how many people do that?) all your getDate() methods return a java.sql.Date object. There is no getCalendar() or getGregorianCalendar() method. You have to get Date objects from your Resultset. Then you have to do your own conversion from one of them to a GregorianCalendar object. Well, GregorianCalendar doesn't have a constructor that accepts java.sql.Date as an argument (why accept a deprecated type as a constructor argument? after all if it was deprecated there is no need to use it, right?). So you have to do getYear(), getMonth() and so on in order to get the values to pass to the GregorianCalendar constructor. But when you do that your compiler will then complain about using deprecated methods. This is retarded.
Now, maybe they fixed it in JDK 1.4. But why did Sun go deprecating methods in 1.1 (which is what the Javadoc says is when getYear() was deprecated) that you can't stop using?
Its these little details that make me react with a bit of hostility toward suggestions about eliminating deprecated methods.
Item 8: It isn't clear to me that turning the primitive data types into full objects is worth whatever advantages that are supposed to accrue from this. It seems to me that Java has bigger problems. I'm mildly resistant to this idea until I see a lot more written pro and con on it.
Item 4: Why ditch the AWT? Swing is pretty much written on top of AWT anyway. AWT is faster too. I don't see what big advantage is supposed to be gained by doing this. AWT exists and Swing relies on it. And why is Mr. Harold once again so keen to break old code? Java's got enough problems as it is without doing that.
Item 3: Yes, the collections have problems. I would suggest that writing a whole new set of collections classes while keeping the old ones is the way to go. The best way to do that is to have an open source group do it under a LGPL or BSD/Apache style license. Create code that Sun doesn't control. The biggest problem with Java now is that Sun is in charge and they just don't have enough bandwidth to improve everything in Java that needs to be improved.
Item 2: Yes, redesign the IO. Just keep the new classes separate from the old classes. Again, better if a lot of this work could be done by anyone but Sun. That way it can be improved more rapidly. One only has to look at the JDOM effort to see what can be accomplished by open source coders operating out from under the thumb of Sun.
What is great about something like JDOM is that it is getting fixed all the time. You don't have to wait for Sun to rev their JVMs. You can even take the same fixes and run them on various versions of the JDKs. On one hand, as many libraries as possible ought to not be part of the core JDK shipment. On the other hand, beeing able to rev libraries independent of Sun's releases is a really good thing IMO. Okay, then Sun should include libraries like JDOM in the install for ease of distribution but make them separately updateable and not controlled or owned by Sun.
Item 1: Yes, I hear that the class loader has serious limitations. I'm sure the JBoss people know how the class loader ought to work. They went en mass and complained about this issue in the Bug Parade. They are the ones most worked up over the class loader limitations. They rant about the class loader on their developer list (well reasoned rants too IMO).
Okay, so these are my reactions to Mr. Harold's suggestions. In my next post I'll address the things about Java that I think are more important for improving its prospects in the marketplace.
I've been making changes to administration form templates for MovableType blogging software. Well, in order to make a change you first have to figure out the existing form and which fields in the HTML show up where in the rendered page. It would be nice to be able to right click over a control and see that control's NAME and other tag fields. Basically, make it easier to find the underlying location in the HTML that is responsible for a given location on the rendered form. Mozilla has a Properties option. But it doesn't tell the control name and it doesn't for instance say that it is a TEXTAREA control. Instead it gives a DOM element name of HTMLInputElement. Not so helpful.
Another thing that would be nice would be traceability back from the HTML to whatever CSS tags are being used to set attributes of the given control or screen area.
There are existing software systems that are HTML based. For instance, the MovableType that I (along with many others) use for blogging is written in Perl and generates HTML forms for its admin interface. The admin forms have a lot of functionality and I think MovableType is fairly good overall. But there are controls on some forms that I wish were a lot richer in functionality.
For instance, I'm typing this post into an HTML TEXTAREA control. If I right click on the control there is no Search option to search for a piece of text in the control. So if I saw a typo in the posted version of this while looking at my blog main page I'd still have to visually scan for it in the editing form. Well, it would be nice to able to override the TEXTAREA control and tell Mozilla (which is the browser I'm using) to use some other functionally richer control for any control on any page that calls out TEXTAREA. Then I'd be able to use a control that had a spell checker, a search facility, and even some macro defineable keys to make it easier to insert common text strings into my posts (eg the skeleton of an A HREF open and close tag pair). One could even add a TEXTAREA control that was a full HTML editor.
One advantage of being able to tell the browser to use some richer version of TEXTAREA is that one doesn't have to have access to the source code of the server app that generates the pages in order to enhance the functionality of the pages.
Of course browsers would need a plug-in facility that allowed users to add replacement versions of common controls. One way to do that which would be portable would be to allow replacement HTML controls to be coded in Java.
General purpose computers are not optimized for any one class of algorithms. In theory if one could lay out transistor logic to exactly perform a chosen algorithm one should be able to achieve orders of magnitude faster performance while at the same time using much less transistor logic. This has been done for sound and other specialist applications. But the cost is high and if you want to make an algorithm change (either to improve the algorithm or to fix bugs) the cost of remanufacturing the chips makes this approach too expensive for most purposes.
However, there is a class of chips called Field Programmable Gate Arrays that, while they may not be as cheap and fast as dedicated logic chips, are still fast enough for many applications. Since they can be programmed (and even reprogrammed) out in the field they are much cheaper to use. Given the large and growing need for biotech and pharmaceutical companies to process thru large quantities of genetic data using some fairly standard algorithms is is not surprising that some companies have introduced dedicated hardware accelerators using FPGAs which come with support for running bioinformatics algorithms..
Time Logic makes hardware accelerators for bioinformatics applications. Time Logic's hardware accelerator technology is built with field programmable gate arrays (FPGAs). Basically, they translate a program's logic into instructions for reprogramming the logic in gates on chips on their accelerator board. They are able to speed up some big genetic data search and comparison algorithms by one to two orders of magnitude. Note that Time Logic claims the amount of genetic data available is doubling every 6 months.
Paracel, a division of Celera Genomics Group, also provides bioinformatics hardware accelerators.
BioIT World has an article about hardware accelerators for bioinformatics problems.
I think it would be fun to develop applications to run on these accelerators. Anyone out there need some contract programming done to get one of these accelerators working for your application?
There are periodic Usenet group debates about programming editors where various people will proclaim with apparently sincere conviction that their preferred editor is the best. In some of these debates I've tried asking some of the believers of various editor faiths if they'd ever tried various other alternatives. Well, no. They had never used Premia (now Starbase?) CodeWright or Visual Slick Edit or Multi-Edit or assorted other editors. Some claimed to have used another editor about 5 or 10 years previously but not its latest version.
In fact, when I start asking around I discover its hard to find people who have used recent versions of two major editors. So its hard to find anyone who can intelligently compare the features of various editors. I use Visual Slick Edit personally. There are things I want it to do that it doesn't do. But I don't know whether any other editor can do all the things I like about Slick plus the things I wish it did (and I'll eventually make a post with a lot of those wishlist items).
Anyway, how to explain the deep irrational devotion that many programmers have for their editors? The Baby Duck Syndrome. Back in the 1950s Konrad Lorenz first demonstrated imprinting of baby ducks on who they think their mother is. Whatever they see at a certain critical early age is what they decide is mother and they will follow it around after that. Well, a lot of programmers have Baby Duck Syndrome and can't help themselves.
You can download Mozilla v1.1 here.
Mozilla is currently my favorite browser. Once it gained the tabbed browsing feature its greater standards compliance as compared to Opera caused me to switch to Moz for most of my browsing. If you haven't tried Mozilla at all or haven't tried it recently I would suggest taking it out for a spin. Its only a 10 meg download for Windows. Builds are available for most other operating systems as well.
Opera v7's release is getting close and it will usher in full DOM support and other improvements in standards support. I believe its a major rewrite of their rendering engine. So if Opera v7 maintains Opera's its advantage in lower memory usage it will be worth a try.
While I haven't tried it BroadPage looks interesting. Its a free browser built on top of IE that supports tabbed browsing. You will find a link to download BroadPage here too.
Also, for tabbed browsing you can check out (and I have not tried) IE plugins that give IE that capability. Surftabs is free and its even open source. Netcaptor costs $29.95 and works only with IE 4 and 5. Surftabs is worth a try IMO.
I had a bug in my Javascript for the Authored On date time update button. The month and day were both too low by 1. The Javascript Date object starts numbering month and day from 0 (like many other date libs - should have known). So if you downloaded in the first few hours you got a bad version. Its fixed now.
In the customize.txt file I've added two new customizations for MovableType.
The first customization makes popup comment boxes resizable for the users. Helps when a lot of users have posted comments and a user wants to read them all.
The second customization is for bloggers. On the Edit Entry form you can add a button that will update the Authored On field to the current date/time on your machine. The Javascript function generates the same date format as MT puts in that control. Give it a try. Great when you do drafts before publishing. MT assigns a date to Authored On the first time a draft is saved. Seems more logical to me to assign the current date/time for when you go to publish. So the button makes that easier to do.
See the text file at this URL for a set of tips for how to customize the administration forms of MovableType for slightly easier blogging. It explains how to make the Main Text Entry control on the Edit Entry form wider and longer and how to make the Title control wider. It also explains how to make the titles in the List and Edit Entries form wider.
If anyone else has customizations that they do to the admin side of MT then send them in and I'll consider them for inclusion in this tips file.
This blog is going to be written for people who spend a lot of time using software tools to do browsing, blogging, editing/compiling/debugging source code, designing databases, and other techie things.
If you just run a blog you will find occasionally useful tips here about how to twiddle MovableType and things to do to improve the way your blog looks and acts. I'll also include some tips about browsing that are useful to bloggers who are out there hunting down and reading thru lots of content.
I have a lot of opinions to offer about what e-mail programs ought to do to make it easier to handle a great deal of e-mail. Of course, one can't care about e-mail without having some pretty strong feelings about Spam and what ought to be done about it. A Monty Python character played by Graham Chapman certainly spoke for me when he/she famously said "I don't want ANY Spam!".
Expect rants about Sun's (mis)handling of Java, Mozilla's development, garish looking web sites which add injury to esthetic revulsion by being hard to read, the fact that too few programmers use best-of-breed editors, and assorted other topics.