I recently wrote a program to compare DNA results, using C#. Basically, it has to compare two CSV files containing about a million rows of data, and spit out a list of matching rows.
On my first attempt, it took 37 seconds to match two files. When I looked into what was taking the most time, it turned out to be my use of RegEx for parsing. Switching to good old Split() and x.Parse() functions brought the time down to 6 seconds. Next, I found that File.ReadAllLines() was taking about 1.5 seconds per file. Switching to File.ReadAllBytes() took only 20 ms, bringing down the time per pair of files to about 2 seconds. Next, I switched to parsing the fields character by character, using mainly switch() statements. In the end, I got the total time to match two files down to 160 ms.
No change of language was required, just tuning effort.
I suspect that this is true with much of our bloated software. It's slow, not because of the language, but because nobody does any real work to improve the performance of software.
It's easy to have an idea about wanting a "better" way to develop software. Many people go farther, actually creating new languages.
Haskell anyone? It's a functional language that's supposed to be superior to classics like C++. It's used by about 0.4% of programmers, according to Stack Overflow. How about Groovy, Scala, Kotlin?
What's hard is to get people on board, excited about your new programming idea.
I don't know about you, but I REMEMBER having to specify disk sectors in the file name on Commodore 64. I REMEMBER having to choose between "extended" memory and "high" memory, and you had to reboot to change the settings. I REMEMBER having to memorize all kinds of commands that you have to type blind in vi. (Oh wait, you still have to do that in vim!) Edlin anyone?
You guys can go back to your "good old days." I don't want them!
Micro-transactions, as suggested by nohup above, would be dwarfed by the transaction fees of "only pennies." How does that make Bitcoin a viable option for micro-transactions?
So how else would you solve the problem of knowing whether or not you can trust someone attempting to initiate a transaction with you, or whether a review is fake?
OK, so you're calling Venezuela's "cryptocurrency" (which never existed) a legitimate use?
If you're worried about payment reversals, you probably ARE doing something criminal, or at least shady. Try doing business with established, reputable businesses!
No. None of your suggested "legitimate" uses are better served by Bitcoin than by traditional currency.
It's easy to sit behind your classroom podium and prognosticate about how long the work week should be. But try running an actual business, and you'll find out just how hard it is to get everything done, even with an 8-hour day!
I hate LinkedIn. I open it out of habit and accept everyone who adds me because I don't know why I wouldn't
There's part of your problem. The idea is that you're supposed to accept invites only from people you actually KNOW! Since you seem to accept all spam, OF COURSE it's going to suck for you!
Take cars, for example. They are highly complex machines. Yes, there's a manual. But when you go rent a car, do you first read the manual? No! Why would you? The controls are easy enough to use that you can figure them out. You only need the manual if you really need to take it apart and fix something.
Microplastics come primarily from wear and tear of plastic articles, such as polyester clothing. If we make roads out of plastic, tires will grind off bits of microplastic, which will wash off the road.
It WILL help with large plastic waste, which can be recycled. But I see this only making the microplastics problem worse.
At a table, which is typically rectangular, it's hard to hear people who aren't right next to you. This makes larger groups splinter into subgroups, as the article suggests.
If the group is standing around a room, like in someone's kitchen, the group can be larger. I've experienced groups of up to 10 that can carry on a single conversation in this type of setting.
Even with people spread further apart, as in the kitchen, it quickly becomes impossible to have a single conversation with 12 or more people.
It's called paid product placement. If Pillsbury wants their products at eye level at your local Kroger, they have to pay Kroger "slotting fees" for that prime placement. If they don't pay, their product goes to the very bottom or very top shelf, where it's hard to see and find.
So if you're looking for better prices, look high and low on the grocery store shelves.
Amazon is doing the exact same thing, but with virtual shelf space.
You mean, an ancient time when different internet services were hosted on different subdomains? Like mail.domain.com or ftp.domain.com? Seems this is still going on all over the Internet. There is, after all, more to the Internet than just the Web!
Good grief. Of course you didn't read the article. They're not changing the actual URL, only HIDING the www part. It's still there for your links and copy/paste.
Probiotics are good, probiotics are not so good. Coffee is bad, coffee is good. Wine is good, wine is bad. Calories are bad, no carbs are bad.
Nutrition is complicated. Science will continue to find seemingly contradictory conclusions forever.
So what's a person to do? Do what Grandma always said: don't overdo anything. Sure, eat yogurt, but don't live on the stuff. Drink coffee, but don't drink ONLY coffee. Drink wine, but not too much. And so on.
Slashdot Mirror
I recently wrote a program to compare DNA results, using C#. Basically, it has to compare two CSV files containing about a million rows of data, and spit out a list of matching rows.
On my first attempt, it took 37 seconds to match two files.
When I looked into what was taking the most time, it turned out to be my use of RegEx for parsing. Switching to good old Split() and x.Parse() functions brought the time down to 6 seconds.
Next, I found that File.ReadAllLines() was taking about 1.5 seconds per file. Switching to File.ReadAllBytes() took only 20 ms, bringing down the time per pair of files to about 2 seconds.
Next, I switched to parsing the fields character by character, using mainly switch() statements. In the end, I got the total time to match two files down to 160 ms.
No change of language was required, just tuning effort.
I suspect that this is true with much of our bloated software. It's slow, not because of the language, but because nobody does any real work to improve the performance of software.
It's easy to have an idea about wanting a "better" way to develop software. Many people go farther, actually creating new languages.
Haskell anyone? It's a functional language that's supposed to be superior to classics like C++. It's used by about 0.4% of programmers, according to Stack Overflow. How about Groovy, Scala, Kotlin?
What's hard is to get people on board, excited about your new programming idea.
I don't know about you, but I REMEMBER having to specify disk sectors in the file name on Commodore 64.
I REMEMBER having to choose between "extended" memory and "high" memory, and you had to reboot to change the settings.
I REMEMBER having to memorize all kinds of commands that you have to type blind in vi. (Oh wait, you still have to do that in vim!)
Edlin anyone?
You guys can go back to your "good old days." I don't want them!
Bitcoin may be king, but that's not what Venezuela went with. They made up their own "cryptocurrency" based on future oil revenues.
Micro-transactions, as suggested by nohup above, would be dwarfed by the transaction fees of "only pennies." How does that make Bitcoin a viable option for micro-transactions?
So how else would you solve the problem of knowing whether or not you can trust someone attempting to initiate a transaction with you, or whether a review is fake?
OK, so you're calling Venezuela's "cryptocurrency" (which never existed) a legitimate use?
If you're worried about payment reversals, you probably ARE doing something criminal, or at least shady. Try doing business with established, reputable businesses!
No. None of your suggested "legitimate" uses are better served by Bitcoin than by traditional currency.
Credit scores have been around for a long time. They aren't perfect, but they do a reasonable job of predicting whether a borrower will default.
Why do we feel Apple's move is creepy, but not TransUnion?
Maybe we could one day set our email inboxes to accept emails only from a sender with a trust score over some threshold. That sounds good to me!
It's easy to sit behind your classroom podium and prognosticate about how long the work week should be. But try running an actual business, and you'll find out just how hard it is to get everything done, even with an 8-hour day!
I hate LinkedIn. I open it out of habit and accept everyone who adds me because I don't know why I wouldn't
There's part of your problem. The idea is that you're supposed to accept invites only from people you actually KNOW! Since you seem to accept all spam, OF COURSE it's going to suck for you!
it wasn't built right.
Take cars, for example. They are highly complex machines. Yes, there's a manual. But when you go rent a car, do you first read the manual? No! Why would you? The controls are easy enough to use that you can figure them out. You only need the manual if you really need to take it apart and fix something.
Software and gadgets shouldn't be any different.
...you haven't built many automation systems.
Automation, whether software or robotic, is HARD to get right.
Even the stuff called "AI" is difficult and expensive to train, and any time you throw something new at it, you have to start over.
Robots ARE taking over all our most mundane, routine jobs, yes. But they will not be able to take over more complex jobs for a long, long time.
Why would this help?
Microplastics come primarily from wear and tear of plastic articles, such as polyester clothing. If we make roads out of plastic, tires will grind off bits of microplastic, which will wash off the road.
It WILL help with large plastic waste, which can be recycled. But I see this only making the microplastics problem worse.
At a table, which is typically rectangular, it's hard to hear people who aren't right next to you. This makes larger groups splinter into subgroups, as the article suggests.
If the group is standing around a room, like in someone's kitchen, the group can be larger. I've experienced groups of up to 10 that can carry on a single conversation in this type of setting.
Even with people spread further apart, as in the kitchen, it quickly becomes impossible to have a single conversation with 12 or more people.
OK so they aren't 100% autonomous, but close enough.
Or...how about "Not a Killer Robot"!
Not a Flamethrower
It's called paid product placement. If Pillsbury wants their products at eye level at your local Kroger, they have to pay Kroger "slotting fees" for that prime placement. If they don't pay, their product goes to the very bottom or very top shelf, where it's hard to see and find.
So if you're looking for better prices, look high and low on the grocery store shelves.
Amazon is doing the exact same thing, but with virtual shelf space.
You mean, an ancient time when different internet services were hosted on different subdomains? Like mail.domain.com or ftp.domain.com? Seems this is still going on all over the Internet. There is, after all, more to the Internet than just the Web!
Good grief. They're not changing the actual URL, only HIDING the www part. It's still there for your links and copy/paste.
Good grief. Of course you didn't read the article. They're not changing the actual URL, only HIDING the www part. It's still there for your links and copy/paste.
They're not changing the actual URL, only HIDING the www part. It's still there for your links and copy/paste.
They're not changing the actual URL, only HIDING the www part.
And if your site depends on www being there to get the content the user expected, you're doing it wrong!
I love the T-shirt going around:
PLUTO
Never Forget
1930-2006
Apparently some people are taking this as more than a joke!
Rogan manages to pick Musk's brain in great detail and in a refreshingly laid-back manner
Yeah, I'll bet he was laid back. That's the influence of the pot he's smoking.
Too bad pot has side effects that aren't so good for business.
Probiotics are good, probiotics are not so good.
Coffee is bad, coffee is good.
Wine is good, wine is bad.
Calories are bad, no carbs are bad.
Nutrition is complicated. Science will continue to find seemingly contradictory conclusions forever.
So what's a person to do? Do what Grandma always said: don't overdo anything. Sure, eat yogurt, but don't live on the stuff. Drink coffee, but don't drink ONLY coffee. Drink wine, but not too much. And so on.
It's not rocket science.