Just check out jxta.org to see exactly this sort of protcol. It lets you find peers, establish peer groups, share data in flexible ways, etc... It's also open source (Apache license), by the way.
I would love to use one of these (especially the higher-end model) for a build server. I seem to remember a parallel make utility for Solaris (one that allows you to use multiple machines to do a build). Anyone know of something similar for Linux?
--JRZ
So, this question came up in an algorithms class at Princeton. To the best of my memory, the slide answering it said:
"What if P = NP?"
"BAD:
- Many computer scientists out of work
GOOD:
- Perfect societal bliss"
The point is essentially this: verifying the value of an answer to an optimization problem (of any kind) is usually easy. But finding a better solution is usually hard (exponential). So, saying "P=NP" is equivalent to saying "Finding an optimal answer is not really harder than verifying if an answer IS optimal." So, finding the optimal design for an aircraft, the optimal routing for a network packet, the optimal anything, is not really that tough. And that wouldn't be such a bad thing for our society (though "perfect bliss" was probably an exaggeration).
But Qt 3.0 has fairly good cross-platform threading and networking libraries, in addition to its GUI stuff. See http://doc.trolltech.com/3.0/network.html for networking docs and http://doc.trolltech.com/3.0/threads.html for threading docs. Since you're already learning the "Qt way," you might as well use it for these features too.
--JRZ
I'd like to build an 8-way xeon server with 64 gigs of RAM and stay within a $900 budget, but it ain't gonna happen.
First off, you want "1U" and "exapandable," which are inherently opposites. 1Us tend to be somewhat custom in design (NOT easy to build yourself from spare parts), and they carry a price premium. I really doubt you could hit that price point (with any level of quality at all) even if you dropped the 1U requirement. For a dual-AMD, Einux's cases alone run over $700!
So, step back and think about what you really need for this application. Why expandability in such a cheap system? By the time you get around to adding another CPU and disk, it might be more cost effective to build a new 1-CPU box from scratch. I would probably go with a single PII at around 933 mHz. Even then, $700 for a 1U will be hard to reach. You probably want to scour Ubid.com and eBay for failed-dot-com loot...
Good luck,
--JRZ
Borland JBuilder
on
Java IDEs?
·
· Score: 3, Informative
JBuilder is still the best that I've found. The Pro edition will run you about $999 and enterprise around $3000 per seat, but it's incredibly usable, fast, powerful, etc.
Most importantly, it has an amazing OpenTools API for customization. Check out codecentral.borland.com and you can find dozens of (usually open-source) plug-ins that really increase the utilty of the IDE.
Oh, and there's a rumor on the JBuilder newsgroups that version 6 will come out at the end of the year. You might want to check into that if you're making a big purchase and at least get a guarantee of a free upgrade (Borland often gives upgrades to people who bought within the last month or two, but after that it's big $$$).
--JRZ
A lot of these long-life laptops are ultralight, ultra-tiny as well (especially the Crusoe ones), and they drive me nuts with their tiny keys. So why don't any mini-laptop keyboards follow the "happy hacking keyboard" model? They manage to be small, while still having big keys, by omitting crap like caps lock, function keys (use fn+1 for f1), windows keys, home, pause, etc. I really don't need any of that crap, just 26 letters, 10 numbers, and half a dozen special (modifier, enter) keys. Maybe some arrows would be nice too, but that definitely does NOT add up to a 101-key board! Don't even get me started on the space wasted by touchpads (as opposed to trackballs)....
Sorry for the rant!
--JRZ
The closest thing to vtune for Linux is probably OProfile: oprofile.sourceforge.net. It uses the same performance-counter-based method as vtune, but it doesn't have a flashy GUI. It can, however, give an annotated dump of source code with performance numbers for each line.
People who say "crypto is irrelevant, because the terrorists only had knives" are missing the point completely. Notice all the talk describing this incident as a "massive intelligence failure?" That's because the terrorists appear to have used crypto to communicate between fifty people for over a year before they got close to any violent acts. If their calls had been intercepted (seeing as some appear to have had long-time, known bin Laden links, they could very well have been monitored), we might have known about this six months ago and stopped it.
Additionally, remember that the US government is limited by their ability to monitor local civillians. The FBI needs a wiretap warrant to conduct such an investigation, although the burden of suspicion is typically a bit lower than a physical search warrant, it still needs to be granted by a judge for each specific case.
That said, I think this legislation is probably a poor idea. There will be so many foreign companies providing escrow-free cryptographic plug-ins that US laws will be irrelevant. In the end, it's likely that only law-abiding citizens would possess the backdoor-enabled crypto software, which could still be compromised by a third party.
If you're interested in general-purpose clustering (i.e. you don't want to re-write all your apps to use MPI), I really suggest checking out Compaq's Single System Image Clustering (SSIC)project. For Linux, this is basically in a pre-Alpha state, but the older, UnixWare-based version was very strong.
They also have a good comparison of clustering technology features on this slide. For now, you need a shared SCSI disk that can run GFS or something similar, but it may be possible to hook in PVFS eventually for low-end stuff.
Basically, SSIC is like MOSIX, but with killer high availability features. If a node goes down (from hardware, OS, or application failure), its workload is seamlessly migrated to another, functioning node. On MOSIX, unfortunately, each process has a "home node." If the home node goes down, the process is dead. SSIC also does load balancing by process migration, and all of that good, high scalability stuff.
Anyways, just give a look, and check out their slideshow...
While you definitely have a good point about applications that lend themselves to multiple-box clusters rather than SMP, I don't think you should make such a blanket statement as:
Any task that is easily parallelized and has low internal communications requirements would run more effectively on multiple servers than on one SMP behemoth.
I have several problems with this generalization. First, parallelizing over multiple servers always adds overhead (in both $$$ and performance) of its own. How are you going to spread a load over multiple web servers? You need a load balancer, either the dedicated (pricey) hardware kind or a standard server converted over to load balancing service (which doesn't get you the greatest speed or scalability in the world). Even in a scientific application that you spread over several boxes, you need some kind of load balancer or traffic cop to get an equitable distribution of work.
Second, let's not forget that two-way (and even four-way, if you were in the Xeon market to begin with) boxes have gotten much cheaper in the past year or two. Most of the important server availability features, like hot swap drives, hot swap power supplies, ECC RAM, 64 bit PCI, etc., are almost impossible to find on 1-way systems these days.
Finally, there's a huge difference between up-front cost and maintenance costs, with the maintenance usually being more expensive. If you double the amount of rack space you need, double the amount of power you need, and put in the effort to keep both systems perfectly in sync, you'll quickly find that you've blown away that little savings you got at the cash register.
But, on the other hand, I agree with that this business of benchmarking web servers with like 8 and 12 CPUs (where things really get into a different pricing league) is a bit silly.
What does XML have to do with a search engine? Are you planning just to grep through a huge XML document? Because you could do the same thing with HTML just as easily...
> We spend more time trying to get Microsoft-
> supported games to run on Linux than we do
> writing games for Linux.
We? Who is this "We" you're talking about? Do you personally spend lots of time trying to get Microsoft-supported games to work on Linux? I sure don't.
Maybe you should replace the word "We" with "a large number of people who are writing applications that they find useful and sharing them for free out of the goodness of their hearts." Then, your post will make perfect sense.
Well, one of many problems is the unwillingness / inability of content providers to charge realistic rates. Partly, this is due to the very low conversion rate of free->paying customers and to the high overall cost of customer acquisition, but it also seems like many providers have their heads in the sand.
Take, for instance, LinuxGram, which I definitely enjoy. I'd even pay $20 a year for it, which is what I pay for Dr. Dobbs. But their subscription price is $100.00 a year, which I find ludicrous. Or look at the They Might Be Giants Unlimited service on eMusic.com. For access to TMBG songs, plus interviews, they want $10 a month! Isn't that a bit shameless for access to content from a single band?
--JRZ
Well, I was going to go off on another "health care, literacy, electricity, food, and telephones before computers" rant. But I'm sure many other people will do that to. So let's pretend that, hypothetically, it was a good idea to invest in PCs for the third world, and let's see how goddamn dumb this is from a technical perspective:
Serious hacking will be needed to get this to work. The Linux playstation port is still in its infancy, and you'd have to create a nice, durable package that included the harddrive, playstation, and monitor. Of course, playstations don't support monitors out of the box...
As we all know, Sony sells these things at a loss in order to make money on game licensing. So they'd be none too happy about this plan, whereas another, traditional vendor might actually be supportive.
There are way better, traditional alternatives out there. A barebones Celeron 400-or-so can be had for about $125 bucks (with case, floppy, no RAM). Like a Playstation, it will need all the peripherals to be added. However, it's less than half the cost, it runs x86 software, and it can use cheap, standard components. Even after adding RAM, a hard disk, keyboard, etc., you can still squeak in under $300, which is cheaper than the PS2 before the necessary additional components.
Using an expensive Playstation CPU for an education PC is a ridiculous waste of money and resources. Much of the PS2 cost goes into ultra-high performance graphics acceleration, which these PC's don't need at all. The cheap Duron/Celeron alternative will provide quite good performance for all the apps that this type of machine will run.
Damnit, people! Can't journalists and policymakers consult with a geek before they spout of ignorantly on technical matters?
Actually, it's entirely unlike JVMs on the market today, because.NET does not include an interpreter. It always compiles the code natively before running. It's more like a TowerJ or JOVE Java environment.
--JRZ
We all knew they had a clause in their Corel investment that allowed them to require a.NET port from the canucks, but it really is just plain sad that they've gone ahead with it.
The CLI is a sophisticated compiler and virtual machine, just like a JVM. Giving it to Corel, a copmany with ZERO compiler/OS/JVM expertise is like telling the people who wrote the AOL client that they should start developing a new operating system.
--JRZ
Lineo/Montavista -- two of the biggest embedded Linux development companies (it is handy to have some folks who actually know a little about Linux when you're doing this sort of project)
Broadcom -- Multi-billion-dollar maker of networking chips (including home DSL stuff) and STBs
Sounds fairly substantial, when you get down to it...
I think I've tried pretty much every IDE/RAD solution out there for Linux, and, for someone with a VB background, I would suggest Kylix as the number 1 option, and a good Java IDE as a backup.
The other tools claim to be "RAD," but they're really just IDEs + GUI form designers. With Kylix/Delphi on the other hand, you can visually design very sophisticated data bindings, network connections, and so on. Borland's JBuilder Pro and Enterprise (not cheap) IDE's are equally cool, with great database components and amazing JavaBeans support. The free "Foundation" version is nice, but it doesn't give you that many of the real RAD features.
I thought BlackAdder was promising, because Python is my favorite language and Qt is my favorite widget set. But it's little more than a thin wrapper around the free QtDesigner, integrated with a not-so-hot editor. Maybe someday in the future it'll be a real IDE, but I don't know. Plus, the Python-Qt binding is a little awkward (you basically have to know C++ to use it).
Yes, but the servers under consideration don't have the US-III. The entry-level Netra isn't even really a serious US-II, but rather a IIe, which was originally designed for embedded applications. In other words, it's the Celeron of the Sparc world, and it doesn't have anywhere near the performance of its big, 8MB-cache bretheren.
You can see SPEC CPU results for a 500 mHz UltraSparc IIe on this page. Yep, that's a base CINT of 165, which is a helluva lot lower than the 307 result of a mere Pentium-III at 700 mHz (results can be seen here). In fact, no Intel chip has ever turned in a performance nearly that bad since the Spec CPU2000 benchmark was created in late 1999. Ouch.
Since these Netras also have IDE drives, you won't be improving performance along that axis either. I'd definitely go with the Intel options as far better bang-per-buck.
Yeah, you're definitely in the groupware world now, man. Source code control systems are just not designed to support complex, rich document sharing by non-technical users. Suites like Lotus Notes/Domino and Exchange, however, are designed for very little else.
Check out Domino.doc, which is a super-enterprise-grade document collaboration platform that does versioning, archiving, searching, approval and workflow, etc. Not sure if it runs on Linux, but the core of Domino certainly does.
And, yes, I know that nobody wants to admit it, but this is really the area where the MS Exchange + Office platform excels (ooh, accidental pun). Where Lotus needs an add-on (.DOC), much of this searching/versioning is built into the core MS software. Of course, then you have to start administering an NT server...
--JRZ
We mostly use a custome OO Java layer over our database to drive our primary web application. But, occasionally, someone wants to crank out a reporting app with Perl. No problem, just load up DBI and whatever CPAN libraries you want. We have another app written in Python that hooks into the DB, equally easy. And, yes, we have our resident PHP fans who insist on using that for quick apps, so they use the mysql layer for their language and have no problem.
And, oh, did I mention that MySQL, which we're using, is free, fast, cross platform, and well tested by thousands of users over the course of many years? It also has tons of freely-available tools (GUIs, web apps, etc.). None of the OODBMSes can touch that (yes, I actually evaluated Zope and Ozone, the two biggest open source alternatives, and they don't come close to what we're looking for).
But, that said, I WISH we could use an OODBMS that was free or inexpensive (this is a nonprofit institution), cross-language (including scripting!), and standards compliant so that we could move to a competitor if we needed to.
In the future, we'll probably move to Java Data Objects (JDO), which provide an object-relational mapping layer over a traditional RDBMS, but without the complexity of full EJB. See Exlab's Castor project for more info.
--JRZ
awww.. damnit, of course you're right about Ritchie vs. Pike. I've been programming and writing about programming languages (for a comp. sci. semester project) all night long, and the names started to swim...
Sorry!
--JRZ
When developers (Pike + friends) needed an efficient, processor-independent language for systems programming, they created C. Later, when the systems got so huge that they needed a new layer of abstraction, they (Stroustrup et al) looked at the problem and came up with C++.
Guido wanted a language with the readability of ABC, but with exceptions, OOP, and extensibity, while Larry Wall obviously needed a Postmodern Extension and Reporting Language. Java's history is similarly tied in with very specific problems (smart devices, then applets).
A programming language is an answer. If you propose to design one without first asking a concrete question (no, it doesn't count if your question is "what would be a really cool language?"), I suggest that you name it "42" for obvious reasons.
That's not necessarily the case. Even though they're using (resaonably cheap) IDE drives, they can still RAID 1-mirror them to prevent loss of data from hard drive failures. They would, however, have to suffer a half hour of downtime to replace the blown disk, but, despite what e-commerce consultants tell you ("if JimsGardenHoseEmporium doesn't get 5-nines availability, it'll lose all its customers!"), most applications could afford to have a 30-minute period of inaccessibility for 1% of their data at a time. The hard thing is desiging a resilient app that can operate well if a portion of its storage just disappears and then reappears sometime later.
--JRZ
Slashdot Mirror
Just check out jxta.org to see exactly this sort of protcol. It lets you find peers, establish peer groups, share data in flexible ways, etc... It's also open source (Apache license), by the way.
I would love to use one of these (especially the higher-end model) for a build server. I seem to remember a parallel make utility for Solaris (one that allows you to use multiple machines to do a build). Anyone know of something similar for Linux?
--JRZ
So, this question came up in an algorithms class at Princeton. To the best of my memory, the slide answering it said:
"What if P = NP?"
"BAD:
- Many computer scientists out of work
GOOD:
- Perfect societal bliss"
The point is essentially this: verifying the value of an answer to an optimization problem (of any kind) is usually easy. But finding a better solution is usually hard (exponential). So, saying "P=NP" is equivalent to saying "Finding an optimal answer is not really harder than verifying if an answer IS optimal." So, finding the optimal design for an aircraft, the optimal routing for a network packet, the optimal anything, is not really that tough. And that wouldn't be such a bad thing for our society (though "perfect bliss" was probably an exaggeration).
But Qt 3.0 has fairly good cross-platform threading and networking libraries, in addition to its GUI stuff. See http://doc.trolltech.com/3.0/network.html for networking docs and http://doc.trolltech.com/3.0/threads.html for threading docs. Since you're already learning the "Qt way," you might as well use it for these features too.
--JRZ
I'd like to build an 8-way xeon server with 64 gigs of RAM and stay within a $900 budget, but it ain't gonna happen.
First off, you want "1U" and "exapandable," which are inherently opposites. 1Us tend to be somewhat custom in design (NOT easy to build yourself from spare parts), and they carry a price premium. I really doubt you could hit that price point (with any level of quality at all) even if you dropped the 1U requirement. For a dual-AMD, Einux's cases alone run over $700!
So, step back and think about what you really need for this application. Why expandability in such a cheap system? By the time you get around to adding another CPU and disk, it might be more cost effective to build a new 1-CPU box from scratch. I would probably go with a single PII at around 933 mHz. Even then, $700 for a 1U will be hard to reach. You probably want to scour Ubid.com and eBay for failed-dot-com loot...
Good luck,
--JRZ
JBuilder is still the best that I've found. The Pro edition will run you about $999 and enterprise around $3000 per seat, but it's incredibly usable, fast, powerful, etc.
Most importantly, it has an amazing OpenTools API for customization. Check out codecentral.borland.com and you can find dozens of (usually open-source) plug-ins that really increase the utilty of the IDE.
Oh, and there's a rumor on the JBuilder newsgroups that version 6 will come out at the end of the year. You might want to check into that if you're making a big purchase and at least get a guarantee of a free upgrade (Borland often gives upgrades to people who bought within the last month or two, but after that it's big $$$).
--JRZ
A lot of these long-life laptops are ultralight, ultra-tiny as well (especially the Crusoe ones), and they drive me nuts with their tiny keys. So why don't any mini-laptop keyboards follow the "happy hacking keyboard" model? They manage to be small, while still having big keys, by omitting crap like caps lock, function keys (use fn+1 for f1), windows keys, home, pause, etc. I really don't need any of that crap, just 26 letters, 10 numbers, and half a dozen special (modifier, enter) keys. Maybe some arrows would be nice too, but that definitely does NOT add up to a 101-key board! Don't even get me started on the space wasted by touchpads (as opposed to trackballs)....
Sorry for the rant!
--JRZ
You might also want to check out KProf at http://kprof.sourceforge.net/. It's a nice GUI for gprof and functioncheck. FunctionCheck is much more detailed than gprof, but it requires re-linking the app. See http://www710.univ-lyon1.fr/~yperret/fnccheck/prof iler.html for more details.
People who say "crypto is irrelevant, because the terrorists only had knives" are missing the point completely. Notice all the talk describing this incident as a "massive intelligence failure?" That's because the terrorists appear to have used crypto to communicate between fifty people for over a year before they got close to any violent acts. If their calls had been intercepted (seeing as some appear to have had long-time, known bin Laden links, they could very well have been monitored), we might have known about this six months ago and stopped it.
Additionally, remember that the US government is limited by their ability to monitor local civillians. The FBI needs a wiretap warrant to conduct such an investigation, although the burden of suspicion is typically a bit lower than a physical search warrant, it still needs to be granted by a judge for each specific case.
That said, I think this legislation is probably a poor idea. There will be so many foreign companies providing escrow-free cryptographic plug-ins that US laws will be irrelevant. In the end, it's likely that only law-abiding citizens would possess the backdoor-enabled crypto software, which could still be compromised by a third party.
If you're interested in general-purpose clustering (i.e. you don't want to re-write all your apps to use MPI), I really suggest checking out Compaq's Single System Image Clustering (SSIC)project. For Linux, this is basically in a pre-Alpha state, but the older, UnixWare-based version was very strong.
They also have a good comparison of clustering technology features on this slide. For now, you need a shared SCSI disk that can run GFS or something similar, but it may be possible to hook in PVFS eventually for low-end stuff.
Basically, SSIC is like MOSIX, but with killer high availability features. If a node goes down (from hardware, OS, or application failure), its workload is seamlessly migrated to another, functioning node. On MOSIX, unfortunately, each process has a "home node." If the home node goes down, the process is dead. SSIC also does load balancing by process migration, and all of that good, high scalability stuff.
Anyways, just give a look, and check out their slideshow...
--JRZ
I have several problems with this generalization. First, parallelizing over multiple servers always adds overhead (in both $$$ and performance) of its own. How are you going to spread a load over multiple web servers? You need a load balancer, either the dedicated (pricey) hardware kind or a standard server converted over to load balancing service (which doesn't get you the greatest speed or scalability in the world). Even in a scientific application that you spread over several boxes, you need some kind of load balancer or traffic cop to get an equitable distribution of work.
Second, let's not forget that two-way (and even four-way, if you were in the Xeon market to begin with) boxes have gotten much cheaper in the past year or two. Most of the important server availability features, like hot swap drives, hot swap power supplies, ECC RAM, 64 bit PCI, etc., are almost impossible to find on 1-way systems these days.
Finally, there's a huge difference between up-front cost and maintenance costs, with the maintenance usually being more expensive. If you double the amount of rack space you need, double the amount of power you need, and put in the effort to keep both systems perfectly in sync, you'll quickly find that you've blown away that little savings you got at the cash register.
But, on the other hand, I agree with that this business of benchmarking web servers with like 8 and 12 CPUs (where things really get into a different pricing league) is a bit silly.
What does XML have to do with a search engine? Are you planning just to grep through a huge XML document? Because you could do the same thing with HTML just as easily...
> We spend more time trying to get Microsoft-
> supported games to run on Linux than we do
> writing games for Linux.
We? Who is this "We" you're talking about? Do you personally spend lots of time trying to get Microsoft-supported games to work on Linux? I sure don't.
Maybe you should replace the word "We" with "a large number of people who are writing applications that they find useful and sharing them for free out of the goodness of their hearts." Then, your post will make perfect sense.
--JRZ
Well, one of many problems is the unwillingness / inability of content providers to charge realistic rates. Partly, this is due to the very low conversion rate of free->paying customers and to the high overall cost of customer acquisition, but it also seems like many providers have their heads in the sand.
Take, for instance, LinuxGram, which I definitely enjoy. I'd even pay $20 a year for it, which is what I pay for Dr. Dobbs. But their subscription price is $100.00 a year, which I find ludicrous. Or look at the They Might Be Giants Unlimited service on eMusic.com. For access to TMBG songs, plus interviews, they want $10 a month! Isn't that a bit shameless for access to content from a single band?
--JRZ
Damnit, people! Can't journalists and policymakers consult with a geek before they spout of ignorantly on technical matters?
Actually, it's entirely unlike JVMs on the market today, because .NET does not include an interpreter. It always compiles the code natively before running. It's more like a TowerJ or JOVE Java environment.
--JRZ
We all knew they had a clause in their Corel investment that allowed them to require a .NET port from the canucks, but it really is just plain sad that they've gone ahead with it.
The CLI is a sophisticated compiler and virtual machine, just like a JVM. Giving it to Corel, a copmany with ZERO compiler/OS/JVM expertise is like telling the people who wrote the AOL client that they should start developing a new operating system.
--JRZ
Sounds fairly substantial, when you get down to it...
The other tools claim to be "RAD," but they're really just IDEs + GUI form designers. With Kylix/Delphi on the other hand, you can visually design very sophisticated data bindings, network connections, and so on. Borland's JBuilder Pro and Enterprise (not cheap) IDE's are equally cool, with great database components and amazing JavaBeans support. The free "Foundation" version is nice, but it doesn't give you that many of the real RAD features.
I thought BlackAdder was promising, because Python is my favorite language and Qt is my favorite widget set. But it's little more than a thin wrapper around the free QtDesigner, integrated with a not-so-hot editor. Maybe someday in the future it'll be a real IDE, but I don't know. Plus, the Python-Qt binding is a little awkward (you basically have to know C++ to use it).
Anyways, good luck. You might want to check out this page on Kylix for VB refugees if you haven't already. --JRZ
You can see SPEC CPU results for a 500 mHz UltraSparc IIe on this page. Yep, that's a base CINT of 165, which is a helluva lot lower than the 307 result of a mere Pentium-III at 700 mHz (results can be seen here). In fact, no Intel chip has ever turned in a performance nearly that bad since the Spec CPU2000 benchmark was created in late 1999. Ouch.
Since these Netras also have IDE drives, you won't be improving performance along that axis either. I'd definitely go with the Intel options as far better bang-per-buck.
--JRZ
Check out Domino.doc, which is a super-enterprise-grade document collaboration platform that does versioning, archiving, searching, approval and workflow, etc. Not sure if it runs on Linux, but the core of Domino certainly does.
And, yes, I know that nobody wants to admit it, but this is really the area where the MS Exchange + Office platform excels (ooh, accidental pun). Where Lotus needs an add-on (.DOC), much of this searching/versioning is built into the core MS software. Of course, then you have to start administering an NT server... --JRZ
We mostly use a custome OO Java layer over our database to drive our primary web application. But, occasionally, someone wants to crank out a reporting app with Perl. No problem, just load up DBI and whatever CPAN libraries you want. We have another app written in Python that hooks into the DB, equally easy. And, yes, we have our resident PHP fans who insist on using that for quick apps, so they use the mysql layer for their language and have no problem.
And, oh, did I mention that MySQL, which we're using, is free, fast, cross platform, and well tested by thousands of users over the course of many years? It also has tons of freely-available tools (GUIs, web apps, etc.). None of the OODBMSes can touch that (yes, I actually evaluated Zope and Ozone, the two biggest open source alternatives, and they don't come close to what we're looking for).
But, that said, I WISH we could use an OODBMS that was free or inexpensive (this is a nonprofit institution), cross-language (including scripting!), and standards compliant so that we could move to a competitor if we needed to.
In the future, we'll probably move to Java Data Objects (JDO), which provide an object-relational mapping layer over a traditional RDBMS, but without the complexity of full EJB. See Exlab's Castor project for more info.
--JRZ
awww.. damnit, of course you're right about Ritchie vs. Pike. I've been programming and writing about programming languages (for a comp. sci. semester project) all night long, and the names started to swim...
Sorry!
--JRZ
When developers (Pike + friends) needed an efficient, processor-independent language for systems programming, they created C. Later, when the systems got so huge that they needed a new layer of abstraction, they (Stroustrup et al) looked at the problem and came up with C++.
Guido wanted a language with the readability of ABC, but with exceptions, OOP, and extensibity, while Larry Wall obviously needed a Postmodern Extension and Reporting Language. Java's history is similarly tied in with very specific problems (smart devices, then applets).
A programming language is an answer. If you propose to design one without first asking a concrete question (no, it doesn't count if your question is "what would be a really cool language?"), I suggest that you name it "42" for obvious reasons.
--JRZ
That's not necessarily the case. Even though they're using (resaonably cheap) IDE drives, they can still RAID 1-mirror them to prevent loss of data from hard drive failures. They would, however, have to suffer a half hour of downtime to replace the blown disk, but, despite what e-commerce consultants tell you ("if JimsGardenHoseEmporium doesn't get 5-nines availability, it'll lose all its customers!"), most applications could afford to have a 30-minute period of inaccessibility for 1% of their data at a time. The hard thing is desiging a resilient app that can operate well if a portion of its storage just disappears and then reappears sometime later.
--JRZ