Also, i have no idea of what you are talking about. To scale up database solutions to huge mean to totally rethink how one design the database. Using convenient stuff like stored procedures and even worse, triggers, in a large clustered system would be really stupid and a total performance killer. However, if one do design systems like that, one would need someone to blame..
So maybe thats why that isn't a very high priority(for continuent)? Things like that generally belong on application servers which don't mind iterating and algorithmisizing(well you go find a better word then..).
Databases are for set-based operations, and that becomes even more obvious the more the data grows.
Also, achieving a guaranteed zero loss of transactions kind of lessen the need from them in the first place..since the most important aspect of a database transaction is not to commit but its ability to lose itself..wouldn't you say?:-)
And you *always* need middleware to scale databases. It just isn't called middleware by salespeople.
Yes, and I've also had Linux do the same thing. It didn't give an error, but no matter how many times I "kill -9"ed it the process never paid attention to the command and carried on churning away. I guess that's the process rather than the OS, but it's still not always "all-powerful root". : Well, to be THAT powerful you have to use other commands, because it's probably because the process is in an interruptible state, relevant post here: http://www.wplug.org/pipermail/wplug/2004-June/022380.html
...could make lots of money by porting to Linux because of one simple reason. If the OS doesn't cost any money, users can put that money into buying licensed applications, like Photoshop. I am not sure how things really are, but according to my experience, most Photoshop users use it in a pretty small scale environment. Now I am not talking about only the paying users, that might be different, but those who run cracked versions.
I think most of those would actually prefer to run legit applications, having support and updates. And now they are closer than ever to be able to do this, since there are acceptable free alternatives(like openoffice, thunderbird) to almost all the other applications that are used by the normal business. The last big hurdle left is Photoshop(and Illustrator, perhaps). Yes, there are other applications left, but none as widely needed.
Oh, many idiots are allowed to drive in Europe as well.:-) The link below is to Euro-ncap, the primary EU organisation for crash testings. How a car fare in these tests are considered a big deal and is an important sales factor.
The difference in design of american and european cars does rather stem from cultural, economical(read gas price), environmental and political reasons, than the level of driver training, even though most drive with a stick in Europe. Which only the really hard-core ones do in the US..:-)
Evolution *is* a theory. Perhaps they should also teach what "theory" means. No. Actually, the existence of evolution is a thoroughly proven fact. Quite like gravity, as some grandparent said. And it's quite easily demonstrated as well, using banana flies, color adjusting birds and so on. And, oh yeah, you know, DNA. In fact, one of it's key features, natural selection, can be observed at least on Saturday nights.
Hence, they should not be teaching "the theory of Evolution" but just evolution, plain and simple. "The theory of evolution" was just the title of something Darwin published some 149 years ago. We now know way more than he did. So teaching the actual content of "the theory of evolution" would actually be quite bad from a scientific standpoint. However, it would still be a hundred times more plausible than any story conjured up by writers payed by people that had vested interest in it's outcome.
But, your interview style appears to dismiss skills that are crucial to the development of solid software, and worse, your approach seems to dismiss those who have those very skills! No, my style doesn't dismiss anything. On the contrary, it allows the applicant to display skills that he/her would never be able to in, say, a bubble sort algorithm. Also, if you have to use it, making a bubble sort algorithm is to easy to be a real test, and only favors those with it in recent memory. The others have to reinvent it. Also, it does not display any other important traits needed in a large development projects.
But, you are right about one thing, I don't care the least what skills they have if they aren't passionate about their work. Because it is when they're not passionate code like the one you described turns up. And who made it? A "experienced and skilled software engineer". And not because of an error, but because it was easier and allowed him/her to get home an hour earlier..
One question I do ask, which seems to be more in line with your style, is "Tell me about something you developed, or a problem that you solved, of which you're particularly proud."
"It's the difference between deep understanding and shallow understanding." I think that I get a much better view of the developers deeper understanding by looking at their code and talking to them. I mean, anyone can make a bubble sort, just not under those circumstances. It is a quite simple algorithm, it is just that doing it during an interview makes the applicant unnecessarily nervous, making mistake that he/she never would in a real work situation. Geniuses can appear as complete idiots in the wrong context. It is not delta force types who can solve simple problem while under fire i am looking for.
"As for code developed on one's "leasure time", that usually belongs to one's current employer anyway." No it doesn't. Usually people don't sign slavery contracts. Some do though. I have never understood why one would accept such idiotic terms. Those people would probably not be very interesting anyway.
"And performance always matters when you scale up" Will you always scale up? Will customers pay for systems that scale well? Not necessarily. Enter pragmatism.
"Tell me, how do you count the number of bits set in a 32-bit word?" Again. You are talking about development that only a minute number of companies do these days.
"I know, you can "subcontract it out". How can you tell if the subcontracter you hire is any good?" Well I guess i'll just have to. Maybe through references?
Donald Knuth knows exactly the state of software development. And somehow I knew that he would be an idol of yours.:-)
"This is the "I'm surrounded by ignorami, and therefore I can be ignorant myself" line of reasoning." Which is perfectly valid from a business viewpoint. One has to be able to adjust. Don't mistake it for an excuse. Not knowing Big O-notation has nothing to do with ignorance, since you would not not need that competence in 99.9% (or more) of business, and even embedded application. If I would need experts in time complexity and code optimization. I would hire a subcontractor.
"Perhaps you need one in a pinch on some small embedded system? Even if not, you should understand the different ways that data can be sorted so you can pick the appropriate implementation for the job." Most likely, both is quicksort (O(n log n) at least or maybe O(n^2) at worst) probably implemented somewhere by someone in the language and less likely the number of items to sort that short (if you have to think about which to choose). But then again, my customer would not expect me to be doing "pinches" in embedded stuff. I would again hire a subcontractor. That would be the best for all involved.
"And, where would you find this code? Most programming is "work for hire". Others own the code one writes. Unless it is GPL, or one is othewise free to share it. Far better to have the candidate write some code and talk about it."
People that are truly passionate about programming, do it on their leisure time as well, that's were I'll find it. Also, this code tend to be more representative of their own style of programming and more probably a result of their own design and review. I don't expect the developer to write ad-hoc code while working for me. This is not an ability that I look for.
""Best" is a vague term" No. It is an exact term. The core competences which you go on needed are the easiest part to verify. However, the first impression you get of someone is seldom correct anyway.
"Win32 cold? MFC? Socket coding? Corba? ATL? " Well, since this is 2008 and I need to make money and compete I would not even list any those technologies. For starters. I would suffice if the developer knew OF those technologies.
"That's like saying, "The only thing that driving a car demonstrates is that one has recentltly taken a driving course." On the contrary, it demonstrates a knowledge of fundAmentals," No it is not. I only demonstrates that the developer can make a fairly basic algorithm. Who cares.
Well, I haven't written a sort algorithm since I went to university, which is something like 13 years ago. Why would I? I haven't used Big O-notation since then either, since most do not understand it. And it is not very often it is either practically applicable or economically viable to put that much effort into detailed code optimization.
I would probably not be able to even make a correct bubble sort algoritm on the first try using pseudocode without the possibility of trying out my code, and I would probably not get the notation right either.
I'd rather look at some code that the applicant has written and thinks reflects his/her development style and then discuss it and related stuff in general. It is quite easy to weed out the ones that only regard programming as a job and are unwilling to learn. The best of the rest are which I try to find.
The only thing your test proves is whether the applicant has recently been taking CS-classes, IMO.
Sun is buying direct influence over millions of websites through owning the database product that powers them. And maybe even more importantly they establish a customer relation with their owners. That's half a sale right there, or at least a direct channel.
I am not saying it comes at a bargain price, though...
They used a telescope in New Mexico, belonging to a college in Michigan, that they controlled over the Net. I can't believe the fantastic things they can to over the Net these days. My mother sent me two packs of kitten litter over the Net last week. The Net is truly a useful Net.
The Net is also a great movie about the Net.
Did I mention I like referring to the tube- and pipe network as "the Net"?
Is there some weird crosspost happening here? Or is the point that the "Intel Employee Caught Running OLPC News Site", talked about in the headline, was pissing in the pool of knowledge?
Good point. The only thing I'm wondering is; what are we discussing here? I have no idea. I thought that would appear from reading the comments. Maybe mushrooms?
Its not likely that people are going to switch to linux because one popular proprietary application runs on linux, OSX, and Windows. I think you're completely wrong. I'd say it's THE application that would tip the scales for a whole bunch of desktop users who then would have no reason not to run Linux anymore. No, GIMP is not an option since learning that is much more expensive than keep using windows, also it is not the de facto standard(which Photoshop and Illustrator are).
Last year, many people(like myself) started using CMS, since free CMS:es like Drupal and Joomla now has reached a level of usability where they save so much developing time that it can't be ignored anymore. In fact, I'd say that not using them would, in most web development projects, be irresponsible and bordering on stupid.
So I predict, that from this year and forward, it will become way harder for consulting firms to charge customers lots of money to, for the millionth time, develop the same GUI:s and content handling mechanisms that most web sites need. An other thing is that so many of these firms so unaware of the threat they are facing, that this will come as a complete shock. "What, did our competitor offer a same or higher level of sofistication for a TENTH of our price?".
just a pain in the ass to sort out and impossible to put in neat little boxes. Yep. Actually it is quite interesting to see the industry trying. Since everybody in it has had well-defined roles and jobs(and aspirations) forever... I mean how much has really changed the last 60 or so years in how a record(which it ALWAYS has been) has been marketed and distributed? Not much. Hence, the industry has solidified, and now, there is a possibility that everything will be completely turned over on it's head and suddenly they will have to move really fast to survive. You know, for Britney and the like, it is all about the money, BIG money...And to them(she, her producers, managers), this is a huge opportunity to earn massively more of it, since the customers won't know the difference. I think that when it comes to the big stars, this is just the beginning, it's enough that a few does it, and it is suddenly an alternative to bring up in negotiations with your label.
And as you said, it will be a boon for everyone else as well.
I also know that it didn't used to cost "hundreds of thousands" necessarily. No, of course you are right, that is a completely insane level if one were talking about the cost of making a single recording for the normal band. I must have thought about the costs of the buying studio equipment then vs now, because people were talking of purchasing computers and gear.
I remember though, in the old days, when a 16 or 24 channel 2 inch recorder(+controller gear and Dolby NR units), a then requirement for any serious mainstream recording studio to compete, could cost from 20-30 000 dollars to much, much more to purchase. And tape costs were actually high enough to constrain the productions. The same equipment would nowadays equal some pretty ordinary multichannel sound cards(ok, maybe not totally ordinary then).
Of course nothing is a guarantee of great sound, mic'ing is still an art, or maybe more of a science... However, now, at least, people can afford to try(and maybe fail).
My point is anyway that what you are talking about is quite large productions(i mean, choirs?), most are not like that, they are 5 weird dudes and a tape recorder that want a better sound than what they got recording using one mic during rehersals. And that whats he talked about, if I understood, TFA correctly.
You're right about the singer. What sucks, sucks even worse when you hear it better. It's a horrible revelation for some bands....always fun to watch....moaaahaa:-)
As much as I used to like the Heads, Mr. Byrne, like most of his ilk, exists in a vacuum. (Not his fault, really) I would actually say the complete reverse, that he knows EXACTLY what he is talking about. Obviously, he's not unaware that people wants to, and already do, sell single songs(or "works", for adequate pretense).
Of course he know the things you talk about. But what he is referring to is recording costs and distribution. When he is saying "almost zero" he is talking about a couple of hundred dollars or a figure that even poor musicians can muster themselves. Compared to the hundreds of thousands it used to cost.
Using a laptop, a cheap mixer( $100), and a couple of acceptable (rent the ones rarely used) microphones, and a little bit of imagination, one can actually make a fair quality recordings of even chamber orchestras and gospel choirs...
Of course, of you want to be "different" by doing huge, hence expensive, one-off recordings and involve lot's of people....sure, you would probably need a bit more money. But that is so obviously not what he is talking about.
end of the day you will have managed to build an expensive data center Well, yeah, but I'd say that a quarter of a million dollars is a totally ridiculus figure anyway.
I have run some storage systems in my day, and that figure would only be reached if you consistently chose the most expensive options, and then thew a LOT of man hours on it.
Especially considering that one probably wouldn't build one datacenter for each movie, but rather one(with 2 co-locations on different places on the globe) for each studio...
Re:Just enough...
on
Where are Wii?
·
· Score: 0, Offtopic
Why would Nintendo want to sell more units than it takes to meet their projected sales for the quarter? Why wouldn't they?
What, you think that they makes less money on each after that mark?
Probably, they have lowered their manufacturing costs quite considerably since the WII:s introduction and now makes huge amounts of money off each unit sold.
More interesting what they will do with all the money they make out of the WII?
It is not only a blessing, too much money CAN produce bad projects also(vista?)..
If the WII 2 (WIII?) is as great as the WII, the will have demonstrated that they just didn't happened to get it right this time.
And then they will become HUGE.
Slashdot Mirror
"PG however does not have any support for scaling".
:-)
Maybe you should go tell skype that.
They intend to scale it to handle 1 billion users.
http://highscalability.com/skype-plans-postgresql-scale-1-billion-users
Also, i have no idea of what you are talking about.
To scale up database solutions to huge mean to totally rethink how one design the database.
Using convenient stuff like stored procedures and even worse, triggers, in a large clustered system would be really stupid and a total performance killer. However, if one do design systems like that, one would need someone to blame..
So maybe thats why that isn't a very high priority(for continuent)? Things like that generally belong on application servers which don't mind iterating and algorithmisizing(well you go find a better word then..).
Databases are for set-based operations, and that becomes even more obvious the more the data grows.
Also, achieving a guaranteed zero loss of transactions kind of lessen the need from them in the first place..since the most important aspect of a database transaction is not to commit but its ability to lose itself..wouldn't you say?
And you *always* need middleware to scale databases. It just isn't called middleware by salespeople.
Well, to be THAT powerful you have to use other commands, because it's probably because the process is in an interruptible state, relevant post here:
http://www.wplug.org/pipermail/wplug/2004-June/022380.html
...could make lots of money by porting to Linux because of one simple reason.
If the OS doesn't cost any money, users can put that money into buying licensed applications, like Photoshop.
I am not sure how things really are, but according to my experience, most Photoshop users use it in a pretty small scale environment.
Now I am not talking about only the paying users, that might be different, but those who run cracked versions.
I think most of those would actually prefer to run legit applications, having support and updates.
And now they are closer than ever to be able to do this, since there are acceptable free alternatives(like openoffice, thunderbird) to almost all the other applications that are used by the normal business.
The last big hurdle left is Photoshop(and Illustrator, perhaps). Yes, there are other applications left, but none as widely needed.
Oh, many idiots are allowed to drive in Europe as well. :-)
:-)
The link below is to Euro-ncap, the primary EU organisation for crash testings. How a car fare in these tests are considered a big deal and is an important sales factor.
www.euroncap.com.
The difference in design of american and european cars does rather stem from cultural, economical(read gas price), environmental and political reasons, than the level of driver training, even though most drive with a stick in Europe. Which only the really hard-core ones do in the US..
And it's quite easily demonstrated as well, using banana flies, color adjusting birds and so on. And, oh yeah, you know, DNA.
In fact, one of it's key features, natural selection, can be observed at least on Saturday nights.
Hence, they should not be teaching "the theory of Evolution" but just evolution, plain and simple.
"The theory of evolution" was just the title of something Darwin published some 149 years ago.
We now know way more than he did. So teaching the actual content of "the theory of evolution" would actually be quite bad from a scientific standpoint. However, it would still be a hundred times more plausible than any story conjured up by writers payed by people that had vested interest in it's outcome.
But, you are right about one thing, I don't care the least what skills they have if they aren't passionate about their work.
Because it is when they're not passionate code like the one you described turns up. And who made it? A "experienced and skilled software engineer". And not because of an error, but because it was easier and allowed him/her to get home an hour earlier..
One question I do ask, which seems to be more in line with your style, is "Tell me about something you developed, or a problem that you solved, of which you're particularly proud."
:-)
What, are you hiring?
"It's the difference between deep understanding and shallow understanding."
:-)
I think that I get a much better view of the developers deeper understanding by looking at their code and talking to them.
I mean, anyone can make a bubble sort, just not under those circumstances.
It is a quite simple algorithm, it is just that doing it during an interview makes the applicant unnecessarily nervous, making mistake that he/she never would in a real work situation. Geniuses can appear as complete idiots in the wrong context.
It is not delta force types who can solve simple problem while under fire i am looking for.
"As for code developed on one's "leasure time", that usually belongs to one's current employer anyway."
No it doesn't. Usually people don't sign slavery contracts. Some do though. I have never understood why one would accept such idiotic terms. Those people would probably not be very interesting anyway.
"And performance always matters when you scale up"
Will you always scale up? Will customers pay for systems that scale well? Not necessarily. Enter pragmatism.
"Tell me, how do you count the number of bits set in a 32-bit word?"
Again. You are talking about development that only a minute number of companies do these days.
"I know, you can "subcontract it out". How can you tell if the subcontracter you hire is any good?"
Well I guess i'll just have to. Maybe through references?
Donald Knuth knows exactly the state of software development. And somehow I knew that he would be an idol of yours.
"This is the "I'm surrounded by ignorami, and therefore I can be ignorant myself" line of reasoning."
Which is perfectly valid from a business viewpoint. One has to be able to adjust. Don't mistake it for an excuse.
Not knowing Big O-notation has nothing to do with ignorance, since you would not not need that competence in 99.9% (or more) of business, and even embedded application.
If I would need experts in time complexity and code optimization. I would hire a subcontractor.
"Perhaps you need one in a pinch on some small embedded system? Even if not, you should understand the different ways that data can be sorted so you can pick the appropriate implementation for the job."
Most likely, both is quicksort (O(n log n) at least or maybe O(n^2) at worst) probably implemented somewhere by someone in the language and less likely the number of items to sort that short (if you have to think about which to choose).
But then again, my customer would not expect me to be doing "pinches" in embedded stuff. I would again hire a subcontractor. That would be the best for all involved.
"And, where would you find this code? Most programming is "work for hire". Others own the code one writes. Unless it is GPL, or one is othewise free to share it. Far better to have the candidate write some code and talk about it."
People that are truly passionate about programming, do it on their leisure time as well, that's were I'll find it. Also, this code tend to be more representative of their own style of programming and more probably a result of their own design and review.
I don't expect the developer to write ad-hoc code while working for me. This is not an ability that I look for.
""Best" is a vague term"
No. It is an exact term. The core competences which you go on needed are the easiest part to verify.
However, the first impression you get of someone is seldom correct anyway.
"Win32 cold? MFC? Socket coding? Corba? ATL? "
Well, since this is 2008 and I need to make money and compete I would not even list any those technologies. For starters. I would suffice if the developer knew OF those technologies.
"That's like saying, "The only thing that driving a car demonstrates is that one has recentltly taken a driving course."
On the contrary, it demonstrates a knowledge of fundAmentals,"
No it is not. I only demonstrates that the developer can make a fairly basic algorithm. Who cares.
My point is that we all have different needs.
Well, I haven't written a sort algorithm since I went to university, which is something like 13 years ago. Why would I?
I haven't used Big O-notation since then either, since most do not understand it. And it is not very often it is either practically applicable or economically viable to put that much effort into detailed code optimization.
I would probably not be able to even make a correct bubble sort algoritm on the first try using pseudocode without the possibility of trying out my code, and I would probably not get the notation right either.
I'd rather look at some code that the applicant has written and thinks reflects his/her development style and then discuss it and related stuff in general. It is quite easy to weed out the ones that only regard programming as a job and are unwilling to learn. The best of the rest are which I try to find.
The only thing your test proves is whether the applicant has recently been taking CS-classes, IMO.
Sun is buying direct influence over millions of websites through owning the database product that powers them.
And maybe even more importantly they establish a customer relation with their owners.
That's half a sale right there, or at least a direct channel.
I am not saying it comes at a bargain price, though...
My mother sent me two packs of kitten litter over the Net last week.
The Net is truly a useful Net.
The Net is also a great movie about the Net.
Did I mention I like referring to the tube- and pipe network as "the Net"?
Is there some weird crosspost happening here?
Or is the point that the "Intel Employee Caught Running OLPC News Site", talked about in the headline, was pissing in the pool of knowledge?
Maybe mushrooms?
Last year, many people(like myself) started using CMS, since free CMS:es like Drupal and Joomla now has reached a level of usability where they save so much developing time that it can't be ignored anymore. In fact, I'd say that not using them would, in most web development projects, be irresponsible and bordering on stupid.
So I predict, that from this year and forward, it will become way harder for consulting firms to charge customers lots of money to, for the millionth time, develop the same GUI:s and content handling mechanisms that most web sites need.
An other thing is that so many of these firms so unaware of the threat they are facing, that this will come as a complete shock.
"What, did our competitor offer a same or higher level of sofistication for a TENTH of our price?".
Happy new year to you too!
Hence, the industry has solidified, and now, there is a possibility that everything will be completely turned over on it's head and suddenly they will have to move really fast to survive.
You know, for Britney and the like, it is all about the money, BIG money...And to them(she, her producers, managers), this is a huge opportunity to earn massively more of it, since the customers won't know the difference. I think that when it comes to the big stars, this is just the beginning, it's enough that a few does it, and it is suddenly an alternative to bring up in negotiations with your label.
And as you said, it will be a boon for everyone else as well.
I must have thought about the costs of the buying studio equipment then vs now, because people were talking of purchasing computers and gear.
I remember though, in the old days, when a 16 or 24 channel 2 inch recorder(+controller gear and Dolby NR units), a then requirement for any serious mainstream recording studio to compete, could cost from 20-30 000 dollars to much, much more to purchase. And tape costs were actually high enough to constrain the productions.
The same equipment would nowadays equal some pretty ordinary multichannel sound cards(ok, maybe not totally ordinary then).
Of course nothing is a guarantee of great sound, mic'ing is still an art, or maybe more of a science... However, now, at least, people can afford to try(and maybe fail).
My point is anyway that what you are talking about is quite large productions(i mean, choirs?), most are not like that, they are 5 weird dudes and a tape recorder that want a better sound than what they got recording using one mic during rehersals. And that whats he talked about, if I understood, TFA correctly.
You're right about the singer. What sucks, sucks even worse when you hear it better.
It's a horrible revelation for some bands....always fun to watch....moaaahaa
Of course he know the things you talk about. But what he is referring to is recording costs and distribution.
When he is saying "almost zero" he is talking about a couple of hundred dollars or a figure that even poor musicians can muster themselves. Compared to the hundreds of thousands it used to cost.
Using a laptop, a cheap mixer( $100), and a couple of acceptable (rent the ones rarely used) microphones, and a little bit of imagination, one can actually make a fair quality recordings of even chamber orchestras and gospel choirs...
Of course, of you want to be "different" by doing huge, hence expensive, one-off recordings and involve lot's of people....sure, you would probably need a bit more money.
But that is so obviously not what he is talking about.
Ahaaa it's in the signature..
Does anyone else get have this weird invalid link to a Zonk post as sibling here?