Some seem overly obsessed with machine speed. I know one coder who was so obsessed he made a mess out of the software because it was designed for speed over maintenance, and maintenance fell behind, making for spaghetti code that was even slower than it would have been. Penny-wise, pound foolish.
I like software stacks that can be tuned for shop conventions to remove repetitions grunt-work, but others don't seem to care much about that. Why I am I typing the field title and size in in 4 different places? Why do I have to redefine the schema copy (table structure) if I want one screen or report to show fields in a different order than the other one? Can't I just supply a list of fields to give their order and inclusion that reference the original for the other details?
Few other devs seem to care about this and similar factoring issues with general CRUD idioms, as if it's part of the job. Scaffolding helps when you start out (auto-generating based on the database), but often can't be used when later changing things because customizations have been added that would be overwritten. It's better to have reference-based abstraction than copy engines: A.K.A. auto-bloaters.
I'd rather have regular CRUD be easy so I can focus on domain logic instead of the grunt-work of marshaling rows and fields back and forth. Our stacks turned bicycle science into rocket science and as long as these guys get a paycheck, they don't care. I do care. Dammit Jim, I wanna be an analyst, not a pasta manager!
interns and grads to India to look at the way they ran the businesses. In one memorable example, a clothing company had only a single piece of some major equipment. When it went down the company had everyone do the work by hand. But the cost of only a short amount of down tome would have easily paid for a second backup piece of equipment. But no one was actually doing the cost-benefit analysis. They found people running businesses who just didn't know how to manage money.
There's plenty of similar here also. One boss I had forbid me from putting comments in stored procedures (SQL) because once somebody did that and the compiler mistook the comment for an optimization flag, messing up the results. Therefore, he summarily forbid all comments. All humans of all creeds often get stuck in silly mental ruts.
Also maybe there were side benefits of the employees doing the manufacturing by hand: they learn or refresh their knowledge about how the product is made. A lot of things have "accidental benefits" one doesn't immediate recognize. I find this commonly when automating tasks.
The level of office intrigue, backstabbing, favoritism, and information hoarding that goes on in a typical Asian office is far worse than anything you will see in America.
That's hard to believe because it's pretty strong here. I'm convinced the Dilbert comic strip is a documentary with the names changed. In fact, the author admits he got many ideas from reader mail of actual events. Dilbert was invented in the Good Ol' USA.
I do often hear the structure is much more hierarchical in many parts of Asia: you never question the boss's judgement: the hierarchy is almost absolute. In the USA you are usually permitted to do such politely, sparingly, and subtly.
I was mostly discussing the decision from HR's point of view and/or habit; I didn't claim it was the best approach. They are not IT experts, and often try to fit skills list verbatim to cover their butts in case something goes wrong. It's like a contract: if they can successfully fit the contract (job ad), they've done their job... on paper.
For H1-B generally the hiring company still wants someone above average for the most part
The visa workers I've worked with are a wide variability in quality. The one constant is that they'll often work long hours because they usually have no family, at least not one living in the USA.
If US students have achieved world domination, why are there such a high demand for H1-B Visa's?
Part of the reason is "combo matching". Look at a typical IT job ad: it will have a list of "required" skills, tools, and versions that a particular company happened to pick for themselves. The chance of any one individual matching that list as-is is statistically pretty small.
But if HR can shop the world, the chance of a statistical fit goes up. Whether that's a rational way to pick a tech worker or not is moot, it's the way HR/recruiters typically think.
There are other reasons for the popularity of H1B visas in corporations, but combo-matching is one not directly related to salary, culture, ageism, or politics.
Something tells me a study put together by academics may not match real-world effectiveness. Skills related to teamwork social dynamics, understanding the business domain, and communication often have at least as big an impact as raw academic prowess, especially early in one's career where one has to pretty much shuddup and do what the boss asks.
That's kind of the same thing: you can buy everybody out because you started buying everybody out first. Regardless, it's a positive feedback cycle ("positive" being a type, not a value judgement). Snowball Effect?
It is baffling. It seems that much stuff would require several servers and/or harddrives, and they are not going to all croak at the same time. Thus, a 10% loss seems within realm, but not 100%. That requires a Gold Plated Multi-Layered Fsckup.
Maybe some wayward task script over-wrote them all (inadvertent "worm"), and the compression algorithm that wrote the backups had a big bug that nobody ever noticed until it was too late because they never tested the backups.
Either way, I suspect this is a case of cost-cutting biting them in the giblets.
ACC has used the network effect to get and stay on top. A manager for a graphics department wants to spend as little money as possible on software. ACC has made a one-stop-shop pretty much. You buy/rent ACC and you get the vast majority of what you need to make and manage graphics.
While there are competitors, they are not as complete as ACC, meaning you have to buy and/or learn yet more software to get the missing features. And orgs also don't want a learning curve for newly hired graphic artists. If your shop uses a mish-mash of tools, finding employees who are a ready fit will be harder. Orgs want plug-and-play employees.
It's similar to Microsoft: an org buys Microsoft not because it's the best, but because everybody else knows it, and they cover the gamut of most business needs in a good-enough way. IBM used to occupy that niche, but MS knocked them off the hill.
Slashdot Mirror
That idea was pondered at C2: http://wiki.c2.com/?InstantLan...
It's worked for MS and Oracle for 3+ decades. That's "long term" by corporate standards.
"sleazebag".
There were poor designs in there. A poor design is a poor design.
For example, if you spot a bubble sort algorithm for big datasets, you know its "speed" score is low compared with alternatives without using a timer.
Some seem overly obsessed with machine speed. I know one coder who was so obsessed he made a mess out of the software because it was designed for speed over maintenance, and maintenance fell behind, making for spaghetti code that was even slower than it would have been. Penny-wise, pound foolish.
I like software stacks that can be tuned for shop conventions to remove repetitions grunt-work, but others don't seem to care much about that. Why I am I typing the field title and size in in 4 different places? Why do I have to redefine the schema copy (table structure) if I want one screen or report to show fields in a different order than the other one? Can't I just supply a list of fields to give their order and inclusion that reference the original for the other details?
Few other devs seem to care about this and similar factoring issues with general CRUD idioms, as if it's part of the job. Scaffolding helps when you start out (auto-generating based on the database), but often can't be used when later changing things because customizations have been added that would be overwritten. It's better to have reference-based abstraction than copy engines: A.K.A. auto-bloaters.
I'd rather have regular CRUD be easy so I can focus on domain logic instead of the grunt-work of marshaling rows and fields back and forth. Our stacks turned bicycle science into rocket science and as long as these guys get a paycheck, they don't care. I do care. Dammit Jim, I wanna be an analyst, not a pasta manager!
It can be both: a sad reflection of reality and funny, in a twisted way.
The Presidency is also like that.
Don't worry, the European Union will help out ... oh, wait
A rarity. Bottle him now!
You just jinxed Canada. Thanks a lot.
There's plenty of similar here also. One boss I had forbid me from putting comments in stored procedures (SQL) because once somebody did that and the compiler mistook the comment for an optimization flag, messing up the results. Therefore, he summarily forbid all comments. All humans of all creeds often get stuck in silly mental ruts.
Also maybe there were side benefits of the employees doing the manufacturing by hand: they learn or refresh their knowledge about how the product is made. A lot of things have "accidental benefits" one doesn't immediate recognize. I find this commonly when automating tasks.
That's hard to believe because it's pretty strong here. I'm convinced the Dilbert comic strip is a documentary with the names changed. In fact, the author admits he got many ideas from reader mail of actual events. Dilbert was invented in the Good Ol' USA.
I do often hear the structure is much more hierarchical in many parts of Asia: you never question the boss's judgement: the hierarchy is almost absolute. In the USA you are usually permitted to do such politely, sparingly, and subtly.
https://www.youtube.com/watch?...
I was mostly discussing the decision from HR's point of view and/or habit; I didn't claim it was the best approach. They are not IT experts, and often try to fit skills list verbatim to cover their butts in case something goes wrong. It's like a contract: if they can successfully fit the contract (job ad), they've done their job ... on paper.
The visa workers I've worked with are a wide variability in quality. The one constant is that they'll often work long hours because they usually have no family, at least not one living in the USA.
So that's Adam Smith's "Comparative Advantage".
Part of the reason is "combo matching". Look at a typical IT job ad: it will have a list of "required" skills, tools, and versions that a particular company happened to pick for themselves. The chance of any one individual matching that list as-is is statistically pretty small.
But if HR can shop the world, the chance of a statistical fit goes up. Whether that's a rational way to pick a tech worker or not is moot, it's the way HR/recruiters typically think.
There are other reasons for the popularity of H1B visas in corporations, but combo-matching is one not directly related to salary, culture, ageism, or politics.
Something tells me a study put together by academics may not match real-world effectiveness. Skills related to teamwork social dynamics, understanding the business domain, and communication often have at least as big an impact as raw academic prowess, especially early in one's career where one has to pretty much shuddup and do what the boss asks.
If they are not a monopoly, they are certainly pounding on the door of Monopoly Mansion. 2018 top-5 ecommerce rankings:
Amazon: 48%
Ebay: 7.2%
Walmart: 4.0%
Apple: 3.9%
Home Depot: 1.6%
Source: https://techcrunch.com/2018/11...
That's kind of the same thing: you can buy everybody out because you started buying everybody out first. Regardless, it's a positive feedback cycle ("positive" being a type, not a value judgement). Snowball Effect?
It is baffling. It seems that much stuff would require several servers and/or harddrives, and they are not going to all croak at the same time. Thus, a 10% loss seems within realm, but not 100%. That requires a Gold Plated Multi-Layered Fsckup.
Maybe some wayward task script over-wrote them all (inadvertent "worm"), and the compression algorithm that wrote the backups had a big bug that nobody ever noticed until it was too late because they never tested the backups.
Either way, I suspect this is a case of cost-cutting biting them in the giblets.
Has anyone asked Putie?
ACC has used the network effect to get and stay on top. A manager for a graphics department wants to spend as little money as possible on software. ACC has made a one-stop-shop pretty much. You buy/rent ACC and you get the vast majority of what you need to make and manage graphics.
While there are competitors, they are not as complete as ACC, meaning you have to buy and/or learn yet more software to get the missing features. And orgs also don't want a learning curve for newly hired graphic artists. If your shop uses a mish-mash of tools, finding employees who are a ready fit will be harder. Orgs want plug-and-play employees.
It's similar to Microsoft: an org buys Microsoft not because it's the best, but because everybody else knows it, and they cover the gamut of most business needs in a good-enough way. IBM used to occupy that niche, but MS knocked them off the hill.
It's a winner-take-most economy. Enjoy.
Who says innovation has to be fast? Society is having indigestion already over it.
https://www.jpl.nasa.gov/news/...
Blasphemy! Lockem up with that Galileo bloke.
But Travelocity charges more for Venus.