As an senior IT guy in a large, multi-national bank, I can see where you're coming from. However, I have to reply to a few of your points:
"The shift to Application Service Providers and services on the web (Microsoft's.NET for example) means that there will be fewer, higher paid deep specialists working for megacorps and a lot of us generalists on the streets."
Nope... can't agree with you there. ASPs (and especially MS.NET) are mere wet dreams for the bulk of corporations out there - especially those with lots of bricks-and-mortar to connect. The ASP model has HUGE flaws (bandwidth requirements, SLAs, data ownership, security admin, etc. etc.) that, today, render them useful for only trivial functions. Trust me - been there in a serious way.
"Globalization. I'm not competing with Chicago and New York any more, but with Singapore and Bangalore. It's just as easy to cut code and FTP it around the world as it is across town."
You'd think this is an issue, but it ain't. Fact is, development is a very social thing and having your developers half a world away from your designers/analysts is tough to manage. Few IT departments can employ such a rigid Waterfall methodology that they can produce firm specs and throw them 'over the wall' to Bangalore. Then there's logistics of design changes, integrated testing, etc, etc. Again, been there in a big way.
IMHO, the biggest threat (??) to corporate IT is the whole dot-com fashionista movement. I see way too many corporate execs getting woodies from reading Wired - and then trying to implement the geek-chique environment in their shops. It don't work, it just looks stupid.
People have been vacationing on my anus (or is it yer anus?) for several years?? Guess that explains why I've been so uncomfortably moody lately. Damn, that itches!
Ok...I'm a parent, I live in Canada, my spouse and I both work and abandon our kids to babysitters before/after school, I censor my kids stuff - videogames, TV shows, etc. Guess how I do it? I turn shit off!!!! I surf with my kids!!! I PLAY with my kids!!!!
So, do I like V-Chip. Nope. It's stupid - about as stupid as Net filtering software. It's an opiate for middle-class parental guilt.
HOWEVER, I think parents who want a V-chip should be able to have one. I can get call display for my phone, so why can't I get 'content-display' for my TV?
Ok... I'm a parent and, no, I wouldn't let my kids go into 'adult' parts of town. But as a metaphor for Internet usage, this is really weak! There's a big difference about controlling where your child is PHYSICALLY versus where your child is ON-LINE. I know when and if my kid leaves the house, but it just takes a few quiet clicks to end up on the wrong side of the Net. Make no mistake, the Net offers up some truly unique parenting dilemas.
Having said all that, I agree with the rest of your post. The Net is NOT that place to let youngsters roam free (and to hell with childless libertarians who spout crap about kids coming to grips with the 'real' world, and "porn never hurt me when I was a BBS whippersnapper", and the like).
For my family, we have a few solutions. We DO NOT use censorware 'cause it doesn't work well and provides a false sense of security. I DO check logs, caches, and such from time to time. I DO surf with my kids from time to time and help them build bookmarks for the sites they regularly visit. And finally, we DO keep the kids PC in a corner of kitchen where the monitor is there for all to see.
Does this work? So far, yes.... but my kids are only 5 and 8 and someday they'll be smarter and curiouser. When that time comes, well I hope I've raised 'em right, Beyond that, I guess I'll be their parent....
First off, MySQL, SQL Server, and the like cannot run large data warehouse-like stuff. Think parallelism, striping, hot failover, and the like. As well, REALLY large transaction-processing systems (banking, airlines, etc.) need the Oracles and DB2s of the the world.
As for websites, I think these large indistrial-grade DBMSs have a place too. Consider a Web app server like iPlanet or Websphere. To support transaction-intensive environments (again...banking, airlines, etc.), the session management requirements are HUGE. I don't think I'd trust MySQL for something like that.
Let me save you all the wasted dollars you might be conned into spending on XP. Every few years some new dumbass social-scientist-cum-geek develops a new way to write systems. They typically fall into 2 groups:
1. The British Bureaucrat with loads of process engineering experience, a clipboard, a stopwatch, and a penchant for 1950's corporate culture. These types gave us Information Engineering and Waterfall Methodology; OR,
2. The faux-Silicon-Valley-geek-hangeron that gave us such tripe as RAD (Rapid Appl'n Development).
So you wonder what a smart-ass like me might know about it all. Having worked on lots of different types of projects using different methodologies, the best advice I have for you is:
- PLAN, PLAN, PLAN. Work with your users and know your requirements (at least as best you can).
- limit your development/project team to a maximum of 6 or 7 people and knock down them cubicle walls! COMMUNICATE PEOPLE! BE A TEAM!
- PROTOTYPE your brains out and show the results to your users/customers. Requirements will change, understanding will increase. You'll all feel better. ITERATE this until everyone is happy!
- Do the initial design as a group - all of you. Decide up front where the critical pieces lie and you all agree that code walkthrus will be mandatory for these pieces.
- Code walkthrus will not be taken personally and will focus on structure, not style. Standards should be checked and observed for just the basic stuff - e.g. naming conventions, etc.
REMEMBER.... 3/4 of the success of any software project is not the methodology or the technology. It's how well the people get along and understand the social rules of engagement. The other 1/4 is in getting the requirements right which, again, all about social interaction between users, developers, etc. Anyone (and I mean ANYONE) who purports to have created a successful methodology can really only offers methods for communicating openly. Ultimately, it's all about humans working together.....
I feel dirty for saying this, but do we NEED yet more ways/places to be connected to the Borg Collective? I mean, for the few hours that comprise an airplane trip, is it that important to be wired into the Net? I can see a day looming when the Type A's of the world will have us all slavishly checking our email 24/7. The world is too dehumanizing and fast-paced as it is. Ted K. - we hardly knew ye!
I would agree that Java on the server is a "good thing". I work for a multi-national bank and we use servlets and JSP to run our on-line banking channels (millions of xact per day, 24/7 uptime, yadda, yadda).
I slightly disagree with your comments on client-side Java, though. Applets suck... hands down. Not because they're not functional, though. There are very useful applet implementations in the world. The big problem, IMHO, is insufficient bandwidth to do anything really useful to a wider audience. When Mon and Dad have broadband, then maybe applets will be appealing.
Java client applications, however, are a different story. Again, the corp I currently work for has employed some very sophisticated Java applications (on WinNT...) and they work very well, technically and from a business perspective. That is, they perform well, are reliable, can share corp-specific utility classes with other Java implementations, and our development staff can contribute to more technology environments (cust channels, admin, Intranet, etc.) than was previously possible.
IMHO, the BIGGEST problem with Java today is that it's largely past the hype-cycle, so the media sees it being "dead" (read: less newsworthy). Look at this way, lowly COBOL is still the language that powers most large corporate environments - but it's "dead" too.
In the meantime, MS has recycled the same ol' COM/DCOM/ActiveX/etc/ garbage again as.NET (wasn't it DNA just a few years back). They're irrelevant...
Ummm.... I'm confused about this. My belief was that JavaScript was just an unfortunately-named technology that has little to do with Java. While we're on the topic, what would REALLY be useful is if JavaScript was implemented in a standard way on all broswers. Wouldn't it be nice if it could, for example, exhibit consistent behaviours across browsers when interacting with Java Applets? Of course, the one may as well wish for a consistent DOM implementation, too....*sign*
Regard....
CrazyLegs (don't ask....)
Slashdot Mirror
As an senior IT guy in a large, multi-national bank, I can see where you're coming from. However, I have to reply to a few of your points:
.NET for example) means that there will be fewer, higher paid deep specialists working for megacorps and a lot of us generalists on the streets."
"The shift to Application Service Providers and services on the web (Microsoft's
Nope... can't agree with you there. ASPs (and especially MS.NET) are mere wet dreams for the bulk of corporations out there - especially those with lots of bricks-and-mortar to connect. The ASP model has HUGE flaws (bandwidth requirements, SLAs, data ownership, security admin, etc. etc.) that, today, render them useful for only trivial functions. Trust me - been there in a serious way.
"Globalization. I'm not competing with Chicago and New York any more, but with Singapore and Bangalore. It's just as easy to cut code and FTP it around the world as it is across town."
You'd think this is an issue, but it ain't. Fact is, development is a very social thing and having your developers half a world away from your designers/analysts is tough to manage. Few IT departments can employ such a rigid Waterfall methodology that they can produce firm specs and throw them 'over the wall' to Bangalore. Then there's logistics of design changes, integrated testing, etc, etc. Again, been there in a big way.
IMHO, the biggest threat (??) to corporate IT is the whole dot-com fashionista movement. I see way too many corporate execs getting woodies from reading Wired - and then trying to implement the geek-chique environment in their shops. It don't work, it just looks stupid.
People have been vacationing on my anus (or is it yer anus?) for several years?? Guess that explains why I've been so uncomfortably moody lately. Damn, that itches!
Ok...I'm a parent, I live in Canada, my spouse and I both work and abandon our kids to babysitters before/after school, I censor my kids stuff - videogames, TV shows, etc. Guess how I do it? I turn shit off!!!! I surf with my kids!!! I PLAY with my kids!!!! So, do I like V-Chip. Nope. It's stupid - about as stupid as Net filtering software. It's an opiate for middle-class parental guilt. HOWEVER, I think parents who want a V-chip should be able to have one. I can get call display for my phone, so why can't I get 'content-display' for my TV?
Ok... I'm a parent and, no, I wouldn't let my kids go into 'adult' parts of town. But as a metaphor for Internet usage, this is really weak! There's a big difference about controlling where your child is PHYSICALLY versus where your child is ON-LINE. I know when and if my kid leaves the house, but it just takes a few quiet clicks to end up on the wrong side of the Net. Make no mistake, the Net offers up some truly unique parenting dilemas.
Having said all that, I agree with the rest of your post. The Net is NOT that place to let youngsters roam free (and to hell with childless libertarians who spout crap about kids coming to grips with the 'real' world, and "porn never hurt me when I was a BBS whippersnapper", and the like).
For my family, we have a few solutions. We DO NOT use censorware 'cause it doesn't work well and provides a false sense of security. I DO check logs, caches, and such from time to time. I DO surf with my kids from time to time and help them build bookmarks for the sites they regularly visit. And finally, we DO keep the kids PC in a corner of kitchen where the monitor is there for all to see.
Does this work? So far, yes.... but my kids are only 5 and 8 and someday they'll be smarter and curiouser. When that time comes, well I hope I've raised 'em right, Beyond that, I guess I'll be their parent....
- rant over -
First off, MySQL, SQL Server, and the like cannot run large data warehouse-like stuff. Think parallelism, striping, hot failover, and the like. As well, REALLY large transaction-processing systems (banking, airlines, etc.) need the Oracles and DB2s of the the world.
As for websites, I think these large indistrial-grade DBMSs have a place too. Consider a Web app server like iPlanet or Websphere. To support transaction-intensive environments (again...banking, airlines, etc.), the session management requirements are HUGE. I don't think I'd trust MySQL for something like that.
Am I right or am I just ignorant?
Let me save you all the wasted dollars you might be conned into spending on XP. Every few years some new dumbass social-scientist-cum-geek develops a new way to write systems. They typically fall into 2 groups:
1. The British Bureaucrat with loads of process engineering experience, a clipboard, a stopwatch, and a penchant for 1950's corporate culture. These types gave us Information Engineering and Waterfall Methodology; OR,
2. The faux-Silicon-Valley-geek-hangeron that gave us such tripe as RAD (Rapid Appl'n Development).
So you wonder what a smart-ass like me might know about it all. Having worked on lots of different types of projects using different methodologies, the best advice I have for you is:
- PLAN, PLAN, PLAN. Work with your users and know your requirements (at least as best you can).
- limit your development/project team to a maximum of 6 or 7 people and knock down them cubicle walls! COMMUNICATE PEOPLE! BE A TEAM!
- PROTOTYPE your brains out and show the results to your users/customers. Requirements will change, understanding will increase. You'll all feel better. ITERATE this until everyone is happy!
- Do the initial design as a group - all of you. Decide up front where the critical pieces lie and you all agree that code walkthrus will be mandatory for these pieces.
- Code walkthrus will not be taken personally and will focus on structure, not style. Standards should be checked and observed for just the basic stuff - e.g. naming conventions, etc.
REMEMBER.... 3/4 of the success of any software project is not the methodology or the technology. It's how well the people get along and understand the social rules of engagement. The other 1/4 is in getting the requirements right which, again, all about social interaction between users, developers, etc. Anyone (and I mean ANYONE) who purports to have created a successful methodology can really only offers methods for communicating openly. Ultimately, it's all about humans working together.....
I feel dirty for saying this, but do we NEED yet more ways/places to be connected to the Borg Collective? I mean, for the few hours that comprise an airplane trip, is it that important to be wired into the Net? I can see a day looming when the Type A's of the world will have us all slavishly checking our email 24/7. The world is too dehumanizing and fast-paced as it is. Ted K. - we hardly knew ye!
CrazyLegs
I would agree that Java on the server is a "good thing". I work for a multi-national bank and we use servlets and JSP to run our on-line banking channels (millions of xact per day, 24/7 uptime, yadda, yadda).
.NET (wasn't it DNA just a few years back). They're irrelevant...
I slightly disagree with your comments on client-side Java, though. Applets suck... hands down. Not because they're not functional, though. There are very useful applet implementations in the world. The big problem, IMHO, is insufficient bandwidth to do anything really useful to a wider audience. When Mon and Dad have broadband, then maybe applets will be appealing.
Java client applications, however, are a different story. Again, the corp I currently work for has employed some very sophisticated Java applications (on WinNT...) and they work very well, technically and from a business perspective. That is, they perform well, are reliable, can share corp-specific utility classes with other Java implementations, and our development staff can contribute to more technology environments (cust channels, admin, Intranet, etc.) than was previously possible.
IMHO, the BIGGEST problem with Java today is that it's largely past the hype-cycle, so the media sees it being "dead" (read: less newsworthy). Look at this way, lowly COBOL is still the language that powers most large corporate environments - but it's "dead" too.
In the meantime, MS has recycled the same ol' COM/DCOM/ActiveX/etc/ garbage again as
Regards.... CrazyLegs
Ummm.... I'm confused about this. My belief was that JavaScript was just an unfortunately-named technology that has little to do with Java. While we're on the topic, what would REALLY be useful is if JavaScript was implemented in a standard way on all broswers. Wouldn't it be nice if it could, for example, exhibit consistent behaviours across browsers when interacting with Java Applets? Of course, the one may as well wish for a consistent DOM implementation, too....*sign* Regard.... CrazyLegs (don't ask....)