OMFG just skimming that "bullet" had me on the floor in hysterics! At first I thought it was an April Fools type article, maybe something from the Onion (esp. w/ the pretty pictures!). They somehow managed to inject every damn SWE buzzword into a single title.
Now I wish someone had posted that article here so we could give it mod points. It'd shoot straight to "5 -Funny".
I only hope this thing doesn't get into the hands of too many PHBs...
Just a conceptual tool, not a silver bullet
on
How Do You Use UML?
·
· Score: 4, Insightful
My first exposure to formal UML was in '96. Took a (employer-sponsored) class, and realized I'd been doing UML already using some newfangled tech called "pencil and paper".
A few years later, I did some work for a major aerospace concern that had brought in an army of UML consultants to completely re-org the way they built their flying machines. They had literally wallpapered an entire floor of an office building with UML diagrams. Needless to say, nothing ever came of their delightful illustrations (other than relieving the corporate coffers of much coin).
UML has unfortunately become an industry with lots of fancy GUI tools to draw pretty pictures,
purporting to magically generate code from committee meetings. Just the sort of thing that keeps PHB's in corner offices, at least until the customer asks to see the working final product.
UML is good as a starting point; it provides a lingua franca to flesh out requirements, and to communicate those requirements to implementors (I'm esp. fond of use case scenarios). However, it doesn't translate well to the often chaotic world of actual implementation,
where performance issues, platform specifics, breakneck delivery schedules, and changing/creeping requirements usually crush the elaborate UML blueprints.
UML is a discipline, rather than a technology. Properly used, it will likely improve your ability ro deliver what the customer wanted. Just don't fall into the trap of assuming its a magic wand that will make the hard parts of implementation disappear.
Ingres was originally intended to compete with the likes of Oracle and MS SQL Server, but never had the power or client base.
Actually, Ingres may have been the first open source relational DBMS. I vaguely recall pawing thru the original UCB sources 14+ years ago (I think there was even a book from Dr. Stonebraker with annotated source code). The commercial version obviously departed from the academic base. And a number of databases actually leveraged off of that original Ingres; a few DBMS's actually began life using QUEL as their query language.
As to the original article, the whole "open source beer" notion seems a mixed metaphor:
Open source != Free beer (e.g., MySQL)
Free software != Open source (e.g., eval or developer versions)
A distinction should also be drawn between limited functionality DBMS's (MySQL, etc.) vs. enterprise level decision support systems (Oracle, DB2, et al). Open sourcing the latter would be giving away a lot of IP/competitive advantage that took many years of development to optimize.
Agreed. Leaving the comfy confines of R&D design/coding work, and dropping into the real world using the s/w you and your R&D cohorts put together, is an eye opener, a humbling experience, and very inspirational all at once.
You'll quickly learn how little you really know about the s/w you helped produce, and at the same time you'll get lots of ideas how to make it better. Esp. working with people who don't give
a damn about the elegant architecture you and your coworkers argued and debated about for so long, or those elaborate features you spent man-years building that the end users hardly ever/never use. They just need to keep their widget manufacturing business running.
Sometimes, dogfood is good food (as in eating one's own...)
The focus in this discussion thus far seems to be on individual efforts/exploits (perhaps/.ers should get away from the cubes/keyboards once in awhile ?) and assumes only the nerds qualify.
Want to get the non-nerds involved ? Want them to confront technology in something other than killing orcs and startroopers with a joystick ?
Start in shop class (at least, for those schools still offering shop class)
Give them a few MindStorm kits to play with for a week (There must be a few still available on ebay).
Send in the programmer nerds.
Stand back and let the fun begin.
FWIW: I've been out of HS quite awhile. I built my ELF II as a HS junior. But I had a little different background than many of nerd-dom: a family construction business. I was managing construction sites by the time I was 17. Its called applied knowledge of analytical and communication skills (aka how to play and work well with others), and all too often the young people emerging from high school are woefully lacking in that particular skill.
Contrary to popular/. belief, technology is not solely the domain of coding nerds.
Beer has been attacked from all sides of late, whether its MADD or Atkins. Need I remind you that it was "liquid bread" that built the pyamids, and that modern democracy was born in the pewter mugs of beer swilling patriots ?
As a homebrewer, I beg, nay, demand that you love and care for your beer as it is. There are many good homes where your unfiltered beer would be welcomed. If you are either unable or unwilling to support your beer, please, please send it to a good home with someone who will love and cherish your beer without questioning its color or purity.
May I suggest the Hospice of St. Arnolds ?
While SQL is good at organizing data, it can be very limited and inefficient when it comes to research.
Perhaps for most commodity DBMS's, but Teradata's
SQL is pretty powerful, and has been adding a lot more powerful capabilities esp. in the past few years
(Refer to the
Teradata SQL reference manuals for details). I've written some time-series type queries myself against 1 billion+ rows (rowcounts are probably more meaningful for this discussion) that, while reading a bit like "War and Peace", do an admirable and efficient job.
I have no doubt that Walmart (and others) often take extracts from the Teradata warehouse for use with other apps (e.g., SAS)...but in many cases, that process is inherited from a prior Oracle/DB2/etc. environment, and simply hasn't been updated to exploit Teradata's capabilities.
Its always a challenge for those of us schooled on Teradata to explain to those coming from other DBMS environments that you can usually get the job done with properly crafted SQL and the proper index selection.
Actually, its more like go for a long coffee
break, then spend the next 10 weeks collecting
and analyzing the returned result set. Teradata
ain't MySQL, or Oracle. A file scan on the 460 Tbytes distributed across all the CPUs/disks
wouldn't take that long. However, if you toss in
about 10+ left joins on subqueries with range predicates, then you might be able to take a short vacation...
"...Teradata databases can reasonably claim to be to Oracle as Oracle is to MySQL."
Except it takes 8 Teradata DBAs to manage the 460 TBytes, and 23 Oracle DBAs to manage 1 Gig;^)
(Not a slam on Oracle DBAs, but on the ridiculous management burden of Oracle)
When you have 460TB of data, how the hell do you even begin to search it?
With SQL.
Teradata was built to handle processing very large datasets from day 1. 460 Terabytes distributed across a large number of CPUs and disks working in parallel with a robust SQL implementation isn't really the challenge. The hard part is keeping all those disks spinning when you start pushing MTBF limits, handling the thousands of concurrent users all banging away at the data, and the constant streaming of new data into the system in order to support near real-time DSS.
IIRC, common dictionary terms can't be trademarked. Which is why "Windows" can't be trademarked. Because they've existed for quite sometime now, thank you (you know those glass thingies in your boss's office that lets the scenery in, but keeps the bugs and weather
out ?), not because everyone used the term "Windows".
Hence, "Jade" can't be trademarked, as its a common term for a precious gem. However, if you use a stylized logo that was similar to the logo that your antagonists use, then you've got problems.
That being said, I recommend changing your product/project name,
for the reasons given in other posts in this topic, namely, you need something unique and trademarkable. That way
You pop to the top of googles
You get the chance to C/D some other poor bastard that tries to hijack your trademark!
I recently went thru that process, having chosen a non-dictionary term for a product that turned out to pre-exist for someone else's product in a google search . Now when you google/yahoo/msn
search for my new title, guess who pops on top ?
Not everyone can be a lawyer or investment banker, and almost no one can be a rock star or sports hero.
Perhaps you missed the nineties ? We've got lawyers falling out of trees, and the law schools are still flushing them out the doors. The bagboy at the local Safeway was giving me stock picks. Hip hop labels continue to churn out "stars/artists" like an assembly line. And while not everyone can be a sports hero, most everyone *can* be a sports agent or hanger-on. SO, you see, the US is still the land of opportunity!
Hell, we don't need to actually make anything! Where's the ROI in that ? Esp. when there's always reality TV ! Eaten any raw pig genitals lately ?
Perhaps the US will simply become a weekly TV show for the rest of the world. It'd be the best sitcom
ever produced. Rabid attorneys, greasy politicians, shallow pop stars, doped up athletes, Jerry Springer, Geraldo Rivera...
So just sit back, enjoy the ride, and see if your name shows up when the credits roll.
'Scuse my igorance, but is Radio 4 available on
shortwave ?
I often hajj out in the sticks here in the
US of A w/ nothing but my jeep, tent, change o'
drawers, and shortwave radio...and often keep company late nights with BBC Worldwide. Sure
would be nice to have Arthur and Marvin to keep
me company 'round the campfire.
No its not the next Google. Its not even a new concept. When I wrote DBD/DBIx::Chart, people
questioned my sanity...now I'm questioning my sanity for not charging $1K per seat.
There actually are offshore locations that are both
cheap and "Americanized". E.g., the Phillipines.
I recently made a tech support call to order some
software for which I had the SKU code, name, everything. I got someone in India, and spent 20+ minutes trying to understand him.
A couple weeks later, I made another tech support call to get a replacement part for my laptop. I can't be certain, but I'm betting it was a Phillipine call center. The whole process was done in just a few minutes, and was one of the best tech support experiences I've had.
If I'm ever successful enough to need an offshore call center, I'd check the Phillipines first. Americanized culture, trilingual (English, Spanish, and Tagalog!), and actually cheaper than the Indian alternative.
(FWIW: I'm not Phillipino, and have no personal interest in any offshoring operation. I'm not thrilled about losing jobs here in the US, but, as someone who needs occasional tech support, I'd prefer it take as little time as possible and actually be helpful.)
With the pile o' cash, plus
holdings, and lord knows what else they're sitting on, and assuming no significant world events for
a few years (maybe a risky prop), they - even you - have to decide where to expend your energies,
and risk your capital.
With the commoditization of s/w, MSFT, even with world dominance, may have reached the knee in its curve. Lets face it, they aren't very good at "innovation" (The dictionary definition, not the BS they pass around at marketing events). So when you've got more $$ than God, and instantaneous electronic markets to fiddle with that can be controlled by a handful of savvy analysts, why bother trying to compete with people doing it for free, and often better ? Esp when every major government is trying to kill your buzz w/ antitrust litigation ? Shed the costs, setup
a holding company, and let the money do the work.
(Which, AFIAK, is the whole idea behind hedge
funds)
And "Buying back stock", esp the vast amount reportedly on the block, only tightens that position. Bill, Steve and other closely helds
sit tight, wait for the institutionals to
dump for each quarterly profit to pump up their
percieved yields, and then sit back and enjoy the ride.
Maybe even when OSS wins, Bill & Steve still win. Once you've reached the "what the fuck" stage of wealth, you have the luxury of picking your battles.
Outstanding, but perhaps for other reasons, esp wrt the/. community:
MSFT either feels *very* confident about continued revenues generated by new products - which most reports indicate may be 2+ years away - , or is following the guiding light of its predecessors (GM, USX, et al): the company becomes mature/rich enough that it really becomes nothing more than a money changer.
GM makes more money from mortgages financed thru GMAC than it does selling cars. USX reached the point where it made more money from similar financial dealings than from making/selling steel. So it effectively stopped making steel.
So perhaps Bill & Steve (who of course will likely
benefit enormously from the divvies and stock
price hike) have wised up and realized that, with
$60+ billion to play with, they can make more
$$ from stock manipulation and financial deals than from publishing software. Even by sending all their development, test, and support to India/China, they've probably reached the point
where they can bring in more dough by
firing *all* the grunts, and hiring some analysts, accountants, and lawyers to scalp teenies in the hedgefund market.
For us OSS'ers, the possible downside will be if the boys of Redmond decide that patent litigation is as profitable as making markets. Hopefully, they'll decide "What the fuck", and just
leave us all alone to sweat out our next SourceForge project.
Slashdot Mirror
Now I wish someone had posted that article here so we could give it mod points. It'd shoot straight to "5 -Funny".
I only hope this thing doesn't get into the hands of too many PHBs...
A few years later, I did some work for a major aerospace concern that had brought in an army of UML consultants to completely re-org the way they built their flying machines. They had literally wallpapered an entire floor of an office building with UML diagrams. Needless to say, nothing ever came of their delightful illustrations (other than relieving the corporate coffers of much coin).
UML has unfortunately become an industry with lots of fancy GUI tools to draw pretty pictures, purporting to magically generate code from committee meetings. Just the sort of thing that keeps PHB's in corner offices, at least until the customer asks to see the working final product.
UML is good as a starting point; it provides a lingua franca to flesh out requirements, and to communicate those requirements to implementors (I'm esp. fond of use case scenarios). However, it doesn't translate well to the often chaotic world of actual implementation, where performance issues, platform specifics, breakneck delivery schedules, and changing/creeping requirements usually crush the elaborate UML blueprints.
UML is a discipline, rather than a technology. Properly used, it will likely improve your ability ro deliver what the customer wanted. Just don't fall into the trap of assuming its a magic wand that will make the hard parts of implementation disappear.
Actually, Ingres may have been the first open source relational DBMS. I vaguely recall pawing thru the original UCB sources 14+ years ago (I think there was even a book from Dr. Stonebraker with annotated source code). The commercial version obviously departed from the academic base. And a number of databases actually leveraged off of that original Ingres; a few DBMS's actually began life using QUEL as their query language.
As to the original article, the whole "open source beer" notion seems a mixed metaphor:
A distinction should also be drawn between limited functionality DBMS's (MySQL, etc.) vs. enterprise level decision support systems (Oracle, DB2, et al). Open sourcing the latter would be giving away a lot of IP/competitive advantage that took many years of development to optimize.
You'll quickly learn how little you really know about the s/w you helped produce, and at the same time you'll get lots of ideas how to make it better. Esp. working with people who don't give a damn about the elegant architecture you and your coworkers argued and debated about for so long, or those elaborate features you spent man-years building that the end users hardly ever/never use. They just need to keep their widget manufacturing business running.
Sometimes, dogfood is good food (as in eating one's own...)
Want to get the non-nerds involved ? Want them to confront technology in something other than killing orcs and startroopers with a joystick ?
FWIW: I've been out of HS quite awhile. I built my ELF II as a HS junior. But I had a little different background than many of nerd-dom: a family construction business. I was managing construction sites by the time I was 17. Its called applied knowledge of analytical and communication skills (aka how to play and work well with others), and all too often the young people emerging from high school are woefully lacking in that particular skill.
Contrary to popular /. belief, technology is not solely the domain of coding nerds.
Beer has been attacked from all sides of late, whether its MADD or Atkins. Need I remind you that it was "liquid bread" that built the pyamids, and that modern democracy was born in the pewter mugs of beer swilling patriots ?
As a homebrewer, I beg, nay, demand that you love and care for your beer as it is. There are many good homes where your unfiltered beer would be welcomed. If you are either unable or unwilling to support your beer, please, please send it to a good home with someone who will love and cherish your beer without questioning its color or purity. May I suggest the Hospice of St. Arnolds ?
Perhaps for most commodity DBMS's, but Teradata's SQL is pretty powerful, and has been adding a lot more powerful capabilities esp. in the past few years (Refer to the Teradata SQL reference manuals for details). I've written some time-series type queries myself against 1 billion+ rows (rowcounts are probably more meaningful for this discussion) that, while reading a bit like "War and Peace", do an admirable and efficient job.
I have no doubt that Walmart (and others) often take extracts from the Teradata warehouse for use with other apps (e.g., SAS)...but in many cases, that process is inherited from a prior Oracle/DB2/etc. environment, and simply hasn't been updated to exploit Teradata's capabilities. Its always a challenge for those of us schooled on Teradata to explain to those coming from other DBMS environments that you can usually get the job done with properly crafted SQL and the proper index selection.
Actually, its more like go for a long coffee break, then spend the next 10 weeks collecting and analyzing the returned result set. Teradata ain't MySQL, or Oracle. A file scan on the 460 Tbytes distributed across all the CPUs/disks wouldn't take that long. However, if you toss in about 10+ left joins on subqueries with range predicates, then you might be able to take a short vacation...
Except it takes 8 Teradata DBAs to manage the 460 TBytes, and 23 Oracle DBAs to manage 1 Gig ;^)
(Not a slam on Oracle DBAs, but on the ridiculous management burden of Oracle)
Esp. when the cards and disks are replaced by a beer keg and tap!
With SQL.
Teradata was built to handle processing very large datasets from day 1. 460 Terabytes distributed across a large number of CPUs and disks working in parallel with a robust SQL implementation isn't really the challenge. The hard part is keeping all those disks spinning when you start pushing MTBF limits, handling the thousands of concurrent users all banging away at the data, and the constant streaming of new data into the system in order to support near real-time DSS.
For those inclined to know more, check here.
IIRC, common dictionary terms can't be trademarked. Which is why "Windows" can't be trademarked. Because they've existed for quite sometime now, thank you (you know those glass thingies in your boss's office that lets the scenery in, but keeps the bugs and weather out ?), not because everyone used the term "Windows".
Hence, "Jade" can't be trademarked, as its a common term for a precious gem. However, if you use a stylized logo that was similar to the logo that your antagonists use, then you've got problems.
That being said, I recommend changing your product/project name, for the reasons given in other posts in this topic, namely, you need something unique and trademarkable. That way
- You pop to the top of googles
- You get the chance to C/D some other poor bastard that tries to hijack your trademark!
I recently went thru that process, having chosen a non-dictionary term for a product that turned out to pre-exist for someone else's product in a google search . Now when you google/yahoo/msn search for my new title, guess who pops on top ?Hell, we don't need to actually make anything! Where's the ROI in that ? Esp. when there's always reality TV ! Eaten any raw pig genitals lately ?
Perhaps the US will simply become a weekly TV show for the rest of the world. It'd be the best sitcom ever produced. Rabid attorneys, greasy politicians, shallow pop stars, doped up athletes, Jerry Springer, Geraldo Rivera...
So just sit back, enjoy the ride, and see if your name shows up when the credits roll.
I often hajj out in the sticks here in the US of A w/ nothing but my jeep, tent, change o' drawers, and shortwave radio...and often keep company late nights with BBC Worldwide. Sure would be nice to have Arthur and Marvin to keep me company 'round the campfire.
Any info much appreciated.
No its not the next Google. Its not even a new concept. When I wrote DBD/DBIx::Chart, people questioned my sanity...now I'm questioning my sanity for not charging $1K per seat.
I recently made a tech support call to order some software for which I had the SKU code, name, everything. I got someone in India, and spent 20+ minutes trying to understand him.
A couple weeks later, I made another tech support call to get a replacement part for my laptop. I can't be certain, but I'm betting it was a Phillipine call center. The whole process was done in just a few minutes, and was one of the best tech support experiences I've had.
If I'm ever successful enough to need an offshore call center, I'd check the Phillipines first. Americanized culture, trilingual (English, Spanish, and Tagalog!), and actually cheaper than the Indian alternative.
(FWIW: I'm not Phillipino, and have no personal interest in any offshoring operation. I'm not thrilled about losing jobs here in the US, but, as someone who needs occasional tech support, I'd prefer it take as little time as possible and actually be helpful.)
With the pile o' cash, plus holdings, and lord knows what else they're sitting on, and assuming no significant world events for a few years (maybe a risky prop), they - even you - have to decide where to expend your energies, and risk your capital.
With the commoditization of s/w, MSFT, even with world dominance, may have reached the knee in its curve. Lets face it, they aren't very good at "innovation" (The dictionary definition, not the BS they pass around at marketing events). So when you've got more $$ than God, and instantaneous electronic markets to fiddle with that can be controlled by a handful of savvy analysts, why bother trying to compete with people doing it for free, and often better ? Esp when every major government is trying to kill your buzz w/ antitrust litigation ? Shed the costs, setup a holding company, and let the money do the work. (Which, AFIAK, is the whole idea behind hedge funds)
And "Buying back stock", esp the vast amount reportedly on the block, only tightens that position. Bill, Steve and other closely helds sit tight, wait for the institutionals to dump for each quarterly profit to pump up their percieved yields, and then sit back and enjoy the ride.
Maybe even when OSS wins, Bill & Steve still win. Once you've reached the "what the fuck" stage of wealth, you have the luxury of picking your battles.
MSFT either feels *very* confident about continued revenues generated by new products - which most reports indicate may be 2+ years away - , or is following the guiding light of its predecessors (GM, USX, et al): the company becomes mature/rich enough that it really becomes nothing more than a money changer.
GM makes more money from mortgages financed thru GMAC than it does selling cars. USX reached the point where it made more money from similar financial dealings than from making/selling steel. So it effectively stopped making steel.
So perhaps Bill & Steve (who of course will likely benefit enormously from the divvies and stock price hike) have wised up and realized that, with $60+ billion to play with, they can make more $$ from stock manipulation and financial deals than from publishing software. Even by sending all their development, test, and support to India/China, they've probably reached the point where they can bring in more dough by firing *all* the grunts, and hiring some analysts, accountants, and lawyers to scalp teenies in the hedgefund market.
For us OSS'ers, the possible downside will be if the boys of Redmond decide that patent litigation is as profitable as making markets. Hopefully, they'll decide "What the fuck", and just leave us all alone to sweat out our next SourceForge project.