> Well I see no reason why a database server ought to need patching in a situation like this. The DB > servers ought to be on a private, firewalled, network and only the DB should be accessable and > only from trusted systems. This is just better for so many reasons.
I understand the concern about patching database servers - it's a *huge* challenge, and can easily knock your databases out. On the other hand, patching gives you 'defense in depth' - and a much more secure environment than just relying on the hard-outer-shell of a firewall.
And patching doesn't have to be a killer, for example right now I've got a set of mission-critical DB2 databases on AIX. These databases have dev, test, and production environments - so patches can be tested prior to implementation. Additionally, both AIX & DB2 allow patches to be applied and removed very easily. So, if the patch causes problems it is easily to back out of it.
Of course, this kind of patching requires a bit of administrative formality, extra cost for test vs prod environments, and software that lends itself to patching. And ideally software that doesn't have new patches every other day...
But those linux costs are nothing - compared to the oracle licensing costs:
- $40,000 / CPU for base product
- $10,000 / CPU for partitioning
- $10,000 / CPU for RAC
So, even a trivial Oracle cluster is just not going to come in under a quarter million dollars. Saving a few thousand dollars by going from windows to linux isn't going to make any difference at all.
Unless you have a large unix support staff you can leverage, want to diminish security-related patching & vulnerabilities, etc, etc. But those numbers are a little tougher to quantify.
> What about partial indexes? PostgreSQL has support for them.
Don't think so. ROLAP (a relational implementation of OLAP) is generally used to support adhoc analysis - in which almost any column can be used to restrict rows. You don't want to rely on indexes here - you'll need too many of them, it'll slow down loads too much, and it'll double the size of the data.
The only case in which indexes typically have much value here are bitmap indexes for very low cardinality columns.
This was pretty radical stuff back before 1998 - then Peppers & Rogers began pushing it in the CRM space.
Of course, this has nothing to do with CRM products like siebel, vantive, etc. But instead has everything to do with methodology - in this case Relationalship Marketing. In this methodology Peppers & Rogers are the thought leaders:
http://www.1to1.com/home.aspx?ItemID=548
As far as bestbuy identifying their worst customers. That's fine, nothing really that wrong with it. It will probably just result in them *not* sending promotions, sales info, etc to those customers.
> As for graphics, images, documents, etc, I never usually have those associated with the db in the > first place. Keep em on the filesystem with the web tier stuff. Do you really put this stuff in the > DB? Or were you just remarking in general about splitting up files by what they are used for?
Both actually. I was just generalizing for an example. However, as much as I'd prefer to avoid putting files in a database, it is sometimes necessary.
A few examples of where it's necessary include:
- some content managment solutions where multiple text & multi-media components are considered part of a package that goes thru various authorization/promotion stages before being presented to the customer.
- event management solutions (security, system management, etc) in which which logs need to get saved and tied to events or trouble-tickets. These logs can often be huge - thousands of lines of text or whatever.
- document management systems - where hapless souls are forced to build systems in support of '40s technology - storing images of documents (rather than normalized data from documents).
So, yeah back in the day (early 80s) we used to think that a hundred megabyte database was slightly unsuual, are quite expensive to implement in a relational database. These days, ten gbyte databases are common, and I run into 300+ gbytes all the time.
The improvements that postgresql is making will help it out in this space, whether due to better support for multiple raid arrays or as a predessor to parallel io devices.
> I would love to hear move comments about what OLAP features are missing, as it relates to your > requirements. OLAP is a topic which comes up from time to time but real world use is often not > offered.
Sure - grouping sets, rollup, and cube commands. These allow you to create a cube with subtotals of various dimensions in a single pass of the data. Hugely useful for cross-tab reports, olap/reporting tools, etc.
But, as someone else pointed out, data partitioning & parallelism are the key performance features essential to powerful reporting. That is, indexes just don't cut it when you need to crunch a million rows spread throughout another twenty million.
So, grouping sets are handy - but what's *really* needed is partitioning & paralellism.
> For the uninitiated and lazy, is there any compelling reason why that's better than putting the > database files on a RAID and letting the OS split the table across devices?
Sure, you might want to distribute your data across multiple arrays. For example - keep your logs and tempspace on an fast & expensive raid 0+1 array of fast (15k drives). Then put small OLTP stuff on a another raid 0+1 array. Then put your huge graphic images, documents, etc on a much more economical RAID5 array.
I use multiple arrays all the time for performance and economics (in db2 & oracle) - this is cool to see postgres pick itup.
For example the following query will recursively drill-down through parent-child relationships within a single table (col1 is pkey & col2 is ckey where either column can have a value linked to the other column in another row).
WITH parent (pkey, ckey) AS
(SELECT pkey, ckey
FROM hierarchy
WHERE pkey = 'AAA'
UNION ALL
SELECT c.pkey, c.ckey
FROM hierarchy c,
parent p
WHERE p.ckey = c.pkey ) SELECT pkey,
ckey FROM parent;
This isn't the simplest syntax in the world - but isn't too horrid either. It's basically defining a parent row then continually issuing union statements to get the attached children. And it can be fast.
Anyhow, the above along with tons of examples & discussion was taken from the free db2 cookbook page 256 by Graeme Birchall: http://ourworld.compuserve.com/homepage s/graeme_bi rchall/HTM_COOK.HTM
> The main complant I'm seeing is that an SQL-based database doesn't exactly support a tree-based > structure "out of the box". However, isn't using linked tables to make a one-to-many relationships > one of the first things anybody's taught in a basic how-to-make-a-database class?
Well, you don't normally want to have to implement recursion by manually linking tables together - way too cumbersome.
You're better off using the 'out of the box' support for recursion from the database vendors. I use it in db2 all the time, and have used it in oracle as well successfully.
> An iterative SQL call would be needed unless the overall part depth was known at query time.
Both oracle & db2 have very good support for recursion. DB2 in particular can easily handle hierarchies and networks of unknown depth in a single query.
Keep in mind that Fabian Pascal is generally considered a crackpot purist. He's been insisting for years that there is no such thing as a relational database product - since none implement a purely relational model.
However, he hasn't delivered an exaple of one, he hasn't clearly articulated the differences between his vision and the commercial options, and he apparently refuses to acknowledge that some problems in life fail to fit well into the relational model (hierarchies, networks, inheritance, etc).
Much of what he, Celko, and Date complained about were actually responses by vendors to adapt to the real world. They were somewhat successful - and now SQL can be used successfully to solve a far greater set of problems than Pascal has ever admitted exist.
A perfect example of this nonsense is there insistence that good indexing in a 3NF model outperforms denormalized data in a star schema. Sounds great, absolutely doesn't work. Across eighteen years working with relational databases I've never seen their suggestions work. Of course they have a response to this - the vendors should just "make the databases faster" - like it's fucking magic or something.
Of course, this isn't to say that he's wrong about xquery - trying to work with unbalanced networks or hierarchies in which the rules change change throughout the schema causes a few problems.
We already have extensive support for recursion & networks in the more powerful RDMBS (db2 & oracle for sure). But combining that with data structures supporting optional branching, complete lack of declarative constraints, optional rules, etc - sounds like something that will never work well.
And going back to the days in which you have to spend a day writing code against a hierarchical database in order to answer a simple question sucks.
Hmmm, haven't people gotten tired of the xml hype yet?
> This is an excellent point, but is it so wrong to evaluate a candidate's attitude and drive just as > much as their aptitude? In my experience, I've seen better results from hard-working, honest people > than from very bright, unmotivated jerks.
First off - if the primary value of a certification is that it demonstrates motivation - then wouldn't it be a huge waste of time - likely to only attract the unimaginative? I mean really, someone could demonstrate more attitude & drive by starting a company, running a SIG, etc, etc - and at the end of the day have a lot more than a piece of paper to show for their effort.
Secondly, I most of the 'very bright, unmotivated jerks' I run into are just mismanaged. They're working for someone who doesn't understand their skills and abilities and is unconcerned about developing them. I've seen *so* many of these people turned completely around by just getting the good fortunate to work for a real leader.
> In his commentary on Brooks' work. There are a number of issues Willis comments about, including a > 'sneer' at the software rent and memory rent. And other comments on the expensive costs of computers > at that time.
And don't forget that most large systems are still leased today. Just visit IBM, HP, Sun, etc and look at their large 8-way systems. I see these systems leased far more often than purchased outright.
> frame. Nowadays, the central server is good for no more than saving your Pr0n
No, things haven't chanaged that much on many software projects.
Want to develop with real data? It often makes sense to share a development database - that can be designed, populated, and maintained by the dba.
Developing large, complex analytical applications? Is your production destination a massive cluster? Then you'll probably need a development environment that's at least a small cluster. And no - every developer doesn't get their own cluster.
Need to interface with MQSeries, Websphere, a content manager, and a workflow manager? You really don't want to spend the time to get all that crap working on everyone's pc. Once again, you'll be way better off sharing a development server.
the current & archived content of is freely available. you just need to live with the use of their freely-provided client. Don't like the client? well, that might suck. But they're hardly abusing you.
> Too much Java shit. From the very start, that sucker wants ~200 MB of ram
I feel for you - I'm not a huge fan of java. However, keep in mind that WebSphere is a Java application server, it is 2004, and is probably targeted at application servers with 4+ gbytes of memory.
> DB2, too many java processes make my box want to commit suicide thats all I can say.
Then you need to move up to at least a 386;-). Seriously, I run db2 on my laptop with 1 gbyte of memory with no problems. Then again, it is also really targeted at servers with 2-4+ gbytes in memory. Also, keep in mind that the java based install is optional, you can also install without it.
> Next day, I run my mysql prompt. > friendly mysql> greets you. It is good to be back!
While I understand an impatience with awkward installers, the differences in reliability, flexibility, and performance between mysql & db2 more than make up for a little extra time spent upfront. The DB2 development staff does not have a design philosophy of "sweep errors under the carpet" the way that mysql does (please don't ask me to back that up, you can easily find evidence of this attitude in their work-arounds & known issue lists). And as far as db2 features go - db2 had more features in 1982 than mysql does today.
So yeah, mysql is much easier to install than db2. A bike is much less complex than a car too.
> I estimate, based on absolutely no data, that NIHS > costs the US economy billions of dollars.
and on the opposite end of the spectrum you've got "Invented Here Syndrome" (IHS) - in which companies are sure that commercial applications are 1000s of times better than custom-developed ones.
These companies have plenty of CS skills, but due mainly to their inability to run a project have convinced themselves their their programmers suck, and that real products are 'bought' rather than 'built'. They fail to realize that complex applications aren't bought - they're bought & customized, and that customization can be more expensive than simply building the software in the first place.
I estimate, based on over twenty years of systems development across a dozen industries that IHS is costing US companies billions of dollars in excessive project complexity & licensing costs while at the same time depriving them of significant new opportunities.
> I believe the direct intent is to improve MySQL and > JBoss to go head to head with IBM on this one.
head to head? not a chance: while jboss & mysql have their merits, they are not direct competitors to websphere & db2. That's like comparing bikes to motorcycles, cars to trucks, etc, etc.
Head to head would have involved a huge reseller's discount on selling WebLogic & Oracle. This is more like a quick & dirty move to get something into that solutions category.
> There seems to be an assumption that one goes to areas where crime occurs. Alas, crime comes to us, > and with all the psychotic individuals out there, your survival is not guaranteed even if you cooperate > fully with the criminal's demands.
That's true - but then again most neighborhoods are still very safe. I live in a little liberal tourist town right off the highway - and we have almost zero crime. Well, that's not exactly true - cars are occasionally broken into, but that's about it. Most of my neighbors leave their doors unlocked, and the only crime I've experienced in ten years is twice I've had unlocked cars rifled.
But could I encounter a dangerous situation? Sure, though the odds are against a 'mugger'. Probably a lot more likely to get into an argument with a drunk at a bar. So, yes I could get into a dangerous situation even in my peaceful little town. But I don't think that the unlikely change of violence warrants much more than being aware of my surroundings. A gun is overkill, and would just result in most neighbors thinking I was a wacko.
I used to do that back in the USMC - I converted my walllocker combination to base 7 and then put that on tape on the back of the lock. Everyone in the barracks tried it and failed. Meanwhile I had a nicely documented combination. Of course, I suppose I was fairly lucky that nobody simply removed the tape - but the same combination was also in my wallet along with all my pin numbers. Again in base 7...
Remember - children aren't adults, and can't be treated exactly like adults. There's a reason why you baby-proof a house with a toddler, why twelve year olds are not provided with driving licenses, why thirteen year olds cannot sign contracts, and why fifteen year-olds cannot buy liquor.
They'll eventually get there, and part of the parenting responsibility is to help them get there in one piece - ready to make important decisions.
Until then the parent has to do much of it. So, even if the child is willing to play his gameboy all day instead of going to school - we take the gameboy away and take him to school. Oh yes - and we try to get him to choose that path, but he won't always. And when he won't - we override his decision. Likewise, he may prefer to watch eight hours of spongebob to reading "call of the wild". That's perfectly understandable - since reading well takes a bit of effort. But we can shut the tv off (temporarily or permanently), and encourage him to move on to something more beneficial. Hopefully, the need for this kind of an override diminishes over time as the child gets older, more mature, and ready to make their own decisions.
> Do your children ever go over to their friends' houses? If so, I doubt many parents are as > restrictive as you, so your children will certainly be exposed to some of the things you are trying to > keep from them.
Yep, and I don't sweat it at all. We're not trying to completely isolate the kids as much as prevent the worst part of popular culture from defining them. So, we see movies together, they see television on occasion - and it really isn't a big deal.
1. lack of television inhibiting social development: yeah, it could interfere with the development of a little social context. On the other hand, I'd say that the value of yucking it up over "Where's the Beef?"/"Wasssup", or knowing who's playing some game today - is vastly overrated. Much of the best of our culture can be gleaned from books, literature, film, etc. And after dumping broadcast tv twenty-two years ago I have never felt limited by its loss.
2. gameboy vs computer: the portability of the gameboy makes it much more difficult to limit the use.
3. high quality tv & moderation vs elimination: Nah, I'd disagree here - there's almost ZERO high-quality content on television that isn't 33% advertising. You've got PBS and that's about it. The history channel is notoriously inaccurate, most broadcasting is 33% advertising (pushing message: you are what you own, you deserve a break, blah, blah), and even at its best - it's an entirely passive experience - unlike listening to the radio or reading. Additionally, it's much easier to eliminate television completely than to try to restrict to only high-quality programming.
4. Hand-eye coordination: yeah, i'm sure there's some value to that. Probably about 1% of what most gamers thing however. If you believe that it's really critical, then have your child learn how to play the guitar - he'll spend the same amount of effort - and then learn a truly valuable skill along the way.
5. Non-public radio: is about the only diverse and challenging radio out there. Many of the other stations pander to the worst adolescent tendencies, play music from rediculously-limited playlists, and are stuck in a single music category. On public radio the kids can discover ethnic/folk, jazz, blues, classicial, and pop music. Oh yeah - and as far as rock is concerned - when's the last time you heard a commercial rock station play Tom Waits? I mean between all the advertising, and moronic dj comments?
> The bottom line is, children should be exposed to all facets of culture, not just the ones their > parents find intellectually stimulating.
Right - television and commercial radio will only introduce them to the 10% of our mainstream culture that you can pick up in a new york minute. Once they've got the "go girl" "humor-thing" down they're ready to move on. And neither television nor commercial radio is going to take them there. The library, traveling, public radio, and carefully selected films will take them there.
Of course, this is if you want a child to become intellectual, insightful, spiritual & worldly. If on the other hand, you'd like them to be a glad-handing salesman - then by all means educate them on a steady-stream of television. It'll prepare them with all the pop-phrases and commercial sensibilities most useful in selling cars.
The problem isn't entirely schools and teachers. Sure, they can be contributors - but like most problems there are multiple factors.
The single, largest factor is the child's immediate social group. Typically starting with parents, branching out to siblings, then to cousins & friends. If this social group puts no value on an education, does not read, is not curious - then the child is almost guaranteed not to develop much intellectually. Oh sure, there are exceptions, but just that.
And the parents can almost completely compensate for a poor school system if they want, here's how:
1. restrict all non-productive distractions. This includes television, gameboys, and computer games. In my household there is ZERO broadcast television, ZERO non-public radio, ZERO gameboys, and about 2-4 hours of computer games a week. Some folks think this is hard it isn't - you especially realize this when you find that your children never beg for toys around christmas time - they just don't see the commercials.
2. read stories to your children every day. There's a wealth of great children's literature, and I have yet to find a pack of boys that could resist for a moment a reading of Kipling's Rikki-Tikki-Tavi. Once the television is off, once you start reading the good stuff, and there butts will be solidly planted. You can give them paper & pencils to draw with as well. BTW, I'd consider the fun authors to read: Roald Dahl, Kipling, EB White, Grahame, Mary Norton, Sid Fleischman, Elenor Estes, Joan Aiken, Louis Sachar, Walter Brooks, etc. Oh yeah, and if you've waited until your kids are 15 to start this it might not work. Sometimes it does, sometimes it's too late.
3. Provide them books as gifts 4. Fill the house with books 5. Spend time with them at the library every week 6. Help the children find interesting ways to approach homework 7. Encourage good grades (with allowances tied to grades, etc) 8. Pursue your imagination with them: just do things that are fun and interesting that they can learn from: - bulid a trebuchet - travel to a foreign country - every night read a poem - join a story-telling group - just use your imagination I've got two boys that are in the top of their class in a pretty good school system. We never pushed them - we simply read to them. That's all it took. Once their imaginations were engages the rest happened all of its own.
The single biggest reason that most children leave school with a poor education - is probably that their parents assumed that they could simply "out-source" the responsibility of education to an institution. I suppose this is a recursion problem isn't it?
Slashdot Mirror
> Well I see no reason why a database server ought to need patching in a situation like this. The DB
> servers ought to be on a private, firewalled,
network and only the DB should be accessable and
> only from trusted systems. This is just better for so many reasons.
I understand the concern about patching database servers - it's a *huge* challenge, and can easily knock your databases out. On the other hand, patching gives you 'defense in depth' - and a much more secure environment than just relying on the hard-outer-shell of a firewall.
And patching doesn't have to be a killer, for example right now I've got a set of mission-critical DB2 databases on AIX. These databases have dev, test, and production environments - so patches can be tested prior to implementation. Additionally, both AIX & DB2 allow patches to be applied and removed very easily. So, if the patch causes problems it is easily to back out of it.
Of course, this kind of patching requires a bit of administrative formality, extra cost for test vs prod environments, and software that lends itself to patching. And ideally software that doesn't have new patches every other day...
But those linux costs are nothing - compared to the oracle licensing costs:
- $40,000 / CPU for base product
- $10,000 / CPU for partitioning
- $10,000 / CPU for RAC
So, even a trivial Oracle cluster is just not going to come in under a quarter million dollars. Saving a few thousand dollars by going from windows to linux isn't going to make any difference at all.
Unless you have a large unix support staff you can leverage, want to diminish security-related patching & vulnerabilities, etc, etc. But those numbers are a little tougher to quantify.
> What about partial indexes? PostgreSQL has support for them.
Don't think so. ROLAP (a relational implementation of OLAP) is generally used to support adhoc analysis - in which almost any column can be used to restrict rows. You don't want to rely on indexes here - you'll need too many of them, it'll slow down loads too much, and it'll double the size of the data.
The only case in which indexes typically have much value here are bitmap indexes for very low cardinality columns.
This was pretty radical stuff back before 1998 - then Peppers & Rogers began pushing it in the CRM space.
Of course, this has nothing to do with CRM products like siebel, vantive, etc. But instead has everything to do with methodology - in this case Relationalship Marketing. In this methodology Peppers & Rogers are the thought leaders:
http://www.1to1.com/home.aspx?ItemID=548
As far as bestbuy identifying their worst customers. That's fine, nothing really that wrong with it. It will probably just result in them *not* sending promotions, sales info, etc to those customers.
> As for graphics, images, documents, etc, I never usually have those associated with the db in the
> first place. Keep em on the filesystem with the web tier stuff. Do you really put this stuff in the
> DB? Or were you just remarking in general about splitting up files by what they are used for?
Both actually. I was just generalizing for an example. However, as much as I'd prefer to avoid putting files in a database, it is sometimes necessary.
A few examples of where it's necessary include:
- some content managment solutions where multiple text & multi-media components are considered part of a package that goes thru various authorization/promotion stages before being presented to the customer.
- event management solutions (security, system management, etc) in which which logs need to get saved and tied to events or trouble-tickets. These logs can often be huge - thousands of lines of text or whatever.
- document management systems - where hapless souls are forced to build systems in support of '40s technology - storing images of documents (rather than normalized data from documents).
So, yeah back in the day (early 80s) we used to think that a hundred megabyte database was slightly unsuual, are quite expensive to implement in a relational database. These days, ten gbyte databases are common, and I run into 300+ gbytes all the time.
The improvements that postgresql is making will help it out in this space, whether due to better support for multiple raid arrays or as a predessor to parallel io devices.
> I would love to hear move comments about what OLAP features are missing, as it relates to your
> requirements. OLAP is a topic which comes up from time to time but real world use is often not
> offered.
Sure - grouping sets, rollup, and cube commands. These allow you to create a cube with subtotals of various dimensions in a single pass of the data. Hugely useful for cross-tab reports, olap/reporting tools, etc.
But, as someone else pointed out, data partitioning & parallelism are the key performance features essential to powerful reporting. That is, indexes just don't cut it when you need to crunch a million rows spread throughout another twenty million.
So, grouping sets are handy - but what's *really* needed is partitioning & paralellism.
> For the uninitiated and lazy, is there any compelling reason why that's better than putting the
> database files on a RAID and letting the OS split the table across devices?
Sure, you might want to distribute your data across multiple arrays. For example - keep your logs and tempspace on an fast & expensive raid 0+1 array of fast (15k drives). Then put small OLTP stuff on a another raid 0+1 array. Then put your huge graphic images, documents, etc on a much more economical RAID5 array.
I use multiple arrays all the time for performance and economics (in db2 & oracle) - this is cool to see postgres pick itup.
wait...isn't this the only value of closed source? that piracy doesn't happen?
or just write a recursive query.
;
e s/graeme_bi rchall/HTM_COOK.HTM
For example the following query will recursively drill-down through parent-child relationships within a single table (col1 is pkey & col2 is ckey where either column can have a value linked to the other column in another row).
WITH parent (pkey, ckey) AS
(SELECT pkey, ckey
FROM hierarchy
WHERE pkey = 'AAA'
UNION ALL
SELECT c.pkey, c.ckey
FROM hierarchy c,
parent p
WHERE p.ckey = c.pkey )
SELECT pkey,
ckey
FROM parent
This isn't the simplest syntax in the world - but isn't too horrid either. It's basically defining a parent row then continually issuing union statements to get the attached children. And it can be fast.
Anyhow, the above along with tons of examples & discussion was taken from the free db2 cookbook page 256 by Graeme Birchall:
http://ourworld.compuserve.com/homepag
> The main complant I'm seeing is that an SQL-based database doesn't exactly support a tree-based
> structure "out of the box". However, isn't using linked tables to make a one-to-many relationships
> one of the first things anybody's taught in a basic how-to-make-a-database class?
Well, you don't normally want to have to implement recursion by manually linking tables together - way too cumbersome.
You're better off using the 'out of the box' support for recursion from the database vendors. I use it in db2 all the time, and have used it in oracle as well successfully.
> An iterative SQL call would be needed unless the overall part depth was known at query time.
Both oracle & db2 have very good support for recursion. DB2 in particular can easily handle hierarchies and networks of unknown depth in a single query.
Keep in mind that Fabian Pascal is generally considered a crackpot purist. He's been insisting for years that there is no such thing as a relational database product - since none implement a purely relational model.
However, he hasn't delivered an exaple of one, he hasn't clearly articulated the differences between his vision and the commercial options, and he apparently refuses to acknowledge that some problems in life fail to fit well into the relational model (hierarchies, networks, inheritance, etc).
Much of what he, Celko, and Date complained about were actually responses by vendors to adapt to the real world. They were somewhat successful - and now SQL can be used successfully to solve a far greater set of problems than Pascal has ever admitted exist.
A perfect example of this nonsense is there insistence that good indexing in a 3NF model outperforms denormalized data in a star schema. Sounds great, absolutely doesn't work. Across eighteen years working with relational databases I've never seen their suggestions work. Of course they have a response to this - the vendors should just "make the databases faster" - like it's fucking magic or something.
Of course, this isn't to say that he's wrong about xquery - trying to work with unbalanced networks or hierarchies in which the rules change change throughout the schema causes a few problems.
We already have extensive support for recursion & networks in the more powerful RDMBS (db2 & oracle for sure). But combining that with data structures supporting optional branching, complete lack of declarative constraints, optional rules, etc - sounds like something that will never work well.
And going back to the days in which you have to spend a day writing code against a hierarchical database in order to answer a simple question sucks.
Hmmm, haven't people gotten tired of the xml hype yet?
> This is an excellent point, but is it so wrong to evaluate a candidate's attitude and drive just as
> much as their aptitude? In my experience, I've seen better results from hard-working, honest people
> than from very bright, unmotivated jerks.
First off - if the primary value of a certification is that it demonstrates motivation - then wouldn't it be a huge waste of time - likely to only attract the unimaginative? I mean really, someone could demonstrate more attitude & drive by starting a company, running a SIG, etc, etc - and at the end of the day have a lot more than a piece of paper to show for their effort.
Secondly, I most of the 'very bright, unmotivated jerks' I run into are just mismanaged. They're working for someone who doesn't understand their skills and abilities and is unconcerned about developing them. I've seen *so* many of these people turned completely around by just getting the good fortunate to work for a real leader.
> In his commentary on Brooks' work. There are a number of issues Willis comments about, including a > 'sneer' at the software rent and memory rent. And other comments on the expensive costs of computers
> at that time.
And don't forget that most large systems are still leased today. Just visit IBM, HP, Sun, etc and look at their large 8-way systems. I see these systems leased far more often than purchased outright.
> frame. Nowadays, the central server is good for no more than saving your Pr0n
No, things haven't chanaged that much on many software projects.
Want to develop with real data? It often makes sense to share a development database - that can be designed, populated, and maintained by the dba.
Developing large, complex analytical applications? Is your production destination a massive cluster? Then you'll probably need a development environment that's at least a small cluster. And no - every developer doesn't get their own cluster.
Need to interface with MQSeries, Websphere, a content manager, and a workflow manager? You really don't want to spend the time to get all that crap working on everyone's pc. Once again, you'll be way better off sharing a development server.
etc, etc.
the current & archived content of is freely available. you just need to live with the use of their freely-provided client. Don't like the client? well, that might suck. But they're hardly abusing you.
> Too much Java shit. From the very start, that sucker wants ~200 MB of ram
;-). Seriously, I run db2 on my laptop with 1 gbyte of memory with no problems. Then again, it is also really targeted at servers with 2-4+ gbytes in memory. Also, keep in mind that the java based install is optional, you can also install without it.
I feel for you - I'm not a huge fan of java. However, keep in mind that WebSphere is a Java application server, it is 2004, and is probably targeted at application servers with 4+ gbytes of memory.
> DB2, too many java processes make my box want to commit suicide thats all I can say.
Then you need to move up to at least a 386
> Next day, I run my mysql prompt.
> friendly mysql> greets you. It is good to be back!
While I understand an impatience with awkward installers, the differences in reliability, flexibility, and performance between mysql & db2 more than make up for a little extra time spent upfront. The DB2 development staff does not have a design philosophy of "sweep errors under the carpet" the way that mysql does (please don't ask me to back that up, you can easily find evidence of this attitude in their work-arounds & known issue lists). And as far as db2 features go - db2 had more features in 1982 than mysql does today.
So yeah, mysql is much easier to install than db2. A bike is much less complex than a car too.
> I estimate, based on absolutely no data, that NIHS
> costs the US economy billions of dollars.
and on the opposite end of the spectrum you've got "Invented Here Syndrome" (IHS) - in which companies are sure that commercial applications are 1000s of times better than custom-developed ones.
These companies have plenty of CS skills, but due mainly to their inability to run a project have convinced themselves their their programmers suck, and that real products are 'bought' rather than 'built'. They fail to realize that complex applications aren't bought - they're bought & customized, and that customization can be more expensive than simply building the software in the first place.
I estimate, based on over twenty years of systems development across a dozen industries that IHS is costing US companies billions of dollars in excessive project complexity & licensing costs while at the same time depriving them of significant new opportunities.
> I believe the direct intent is to improve MySQL and
> JBoss to go head to head with IBM on this one.
head to head? not a chance: while jboss & mysql have their merits, they are not direct competitors to websphere & db2. That's like comparing bikes to motorcycles, cars to trucks, etc, etc.
Head to head would have involved a huge reseller's discount on selling WebLogic & Oracle. This is more like a quick & dirty move to get something into that solutions category.
> There seems to be an assumption that one goes to areas where crime occurs. Alas, crime comes to us,
> and with all the psychotic individuals out there, your survival is not guaranteed even if you cooperate > fully with the criminal's demands.
That's true - but then again most neighborhoods are still very safe. I live in a little liberal tourist town right off the highway - and we have almost zero crime. Well, that's not exactly true - cars are occasionally broken into, but that's about it. Most of my neighbors leave their doors unlocked, and the only crime I've experienced in ten years is twice I've had unlocked cars rifled.
But could I encounter a dangerous situation? Sure, though the odds are against a 'mugger'. Probably a lot more likely to get into an argument with a drunk at a bar. So, yes I could get into a dangerous situation even in my peaceful little town. But I don't think that the unlikely change of violence warrants much more than being aware of my surroundings. A gun is overkill, and would just result in most neighbors thinking I was a wacko.
I used to do that back in the USMC - I converted my walllocker combination to base 7 and then put that on tape on the back of the lock. Everyone in the barracks tried it and failed. Meanwhile I had a nicely documented combination. Of course, I suppose I was fairly lucky that nobody simply removed the tape - but the same combination was also in my wallet along with all my pin numbers. Again in base 7...
Remember - children aren't adults, and can't be treated exactly like adults. There's a reason why you baby-proof a house with a toddler, why twelve year olds are not provided with driving licenses, why thirteen year olds cannot sign contracts, and why fifteen year-olds cannot buy liquor.
They'll eventually get there, and part of the parenting responsibility is to help them get there in one piece - ready to make important decisions.
Until then the parent has to do much of it. So, even if the child is willing to play his gameboy all day instead of going to school - we take the gameboy away and take him to school. Oh yes - and we try to get him to choose that path, but he won't always. And when he won't - we override his decision. Likewise, he may prefer to watch eight hours of spongebob to reading "call of the wild". That's perfectly understandable - since reading well takes a bit of effort. But we can shut the tv off (temporarily or permanently), and encourage him to move on to something more beneficial. Hopefully, the need for this kind of an override diminishes over time as the child gets older, more mature, and ready to make their own decisions.
> Do your children ever go over to their friends' houses? If so, I doubt many parents are as
> restrictive as you, so your children will certainly be exposed to some of the things you are trying to
> keep from them.
Yep, and I don't sweat it at all. We're not trying to completely isolate the kids as much as prevent the worst part of popular culture from defining them. So, we see movies together, they see television on occasion - and it really isn't a big deal.
Chris,
I'll address these points one at a time:
1. lack of television inhibiting social development: yeah, it could interfere with the development of a little social context. On the other hand, I'd say that the value of yucking it up over "Where's the Beef?"/"Wasssup", or knowing who's playing some game today - is vastly overrated. Much of the best of our culture can be gleaned from books, literature, film, etc. And after dumping broadcast tv twenty-two years ago I have never felt limited by its loss.
2. gameboy vs computer: the portability of the gameboy makes it much more difficult to limit the use.
3. high quality tv & moderation vs elimination: Nah, I'd disagree here - there's almost ZERO high-quality content on television that isn't 33% advertising. You've got PBS and that's about it. The history channel is notoriously inaccurate, most broadcasting is 33% advertising (pushing message: you are what you own, you deserve a break, blah, blah), and even at its best - it's an entirely passive experience - unlike listening to the radio or reading. Additionally, it's much easier to eliminate television completely than to try to restrict to only high-quality programming.
4. Hand-eye coordination: yeah, i'm sure there's some value to that. Probably about 1% of what most gamers thing however. If you believe that it's really critical, then have your child learn how to play the guitar - he'll spend the same amount of effort - and then learn a truly valuable skill along the way.
5. Non-public radio: is about the only diverse and challenging radio out there. Many of the other stations pander to the worst adolescent tendencies, play music from rediculously-limited playlists, and are stuck in a single music category. On public radio the kids can discover ethnic/folk, jazz, blues, classicial, and pop music. Oh yeah - and as far as rock is concerned - when's the last time you heard a commercial rock station play Tom Waits? I mean between all the advertising, and moronic dj comments?
> The bottom line is, children should be exposed to all facets of culture, not just the ones their
> parents find intellectually stimulating.
Right - television and commercial radio will only introduce them to the 10% of our mainstream culture that you can pick up in a new york minute. Once they've got the "go girl" "humor-thing" down they're ready to move on. And neither television nor commercial radio is going to take them there. The library, traveling, public radio, and carefully selected films will take them there.
Of course, this is if you want a child to become intellectual, insightful, spiritual & worldly. If on the other hand, you'd like them to be a glad-handing salesman - then by all means educate them on a steady-stream of television. It'll prepare them with all the pop-phrases and commercial sensibilities most useful in selling cars.
The problem isn't entirely schools and teachers. Sure, they can be contributors - but like most problems there are multiple factors.
The single, largest factor is the child's immediate social group. Typically starting with parents, branching out to siblings, then to cousins & friends. If this social group puts no value on an education, does not read, is not curious - then the child is almost guaranteed not to develop much intellectually. Oh sure, there are exceptions, but just that.
And the parents can almost completely compensate for a poor school system if they want, here's how:
1. restrict all non-productive distractions. This includes television, gameboys, and computer games. In my household there is ZERO broadcast television, ZERO non-public radio, ZERO gameboys, and about 2-4 hours of computer games a week. Some folks think this is hard it isn't - you especially realize this when you find that your children never beg for toys around christmas time - they just don't see the commercials.
2. read stories to your children every day. There's a wealth of great children's literature, and I have yet to find a pack of boys that could resist for a moment a reading of Kipling's Rikki-Tikki-Tavi. Once the television is off, once you start reading the good stuff, and there butts will be solidly planted. You can give them paper & pencils to draw with as well. BTW, I'd consider the fun authors to read: Roald Dahl, Kipling, EB White, Grahame, Mary Norton, Sid Fleischman, Elenor Estes, Joan Aiken, Louis Sachar, Walter Brooks, etc. Oh yeah, and if you've waited until your kids are 15 to start this it might not work. Sometimes it does, sometimes it's too late.
3. Provide them books as gifts
4. Fill the house with books
5. Spend time with them at the library every week
6. Help the children find interesting ways to approach homework
7. Encourage good grades (with allowances tied to grades, etc)
8. Pursue your imagination with them: just do things that are fun and interesting that they can learn from: - bulid a trebuchet - travel to a foreign country - every night read a poem - join a story-telling group - just use your imagination I've got two boys that are in the top of their class in a pretty good school system. We never pushed them - we simply read to them. That's all it took. Once their imaginations were engages the rest happened all of its own.
The single biggest reason that most children leave school with a poor education - is probably that their parents assumed that they could simply "out-source" the responsibility of education to an institution. I suppose this is a recursion problem isn't it?