Lack of inferiority doesn't necessarily imply superiority:-)
Personally I'm currently scrambling to go from a RoR project to a horrible C/BASIC/Canvas based ETL *thingy*. As always it's more about the project/how long you've been on it/what stage it's in than the specific technology.
Hey, maybe you're right. Maybe we've reached the endgame as far as computing abstractions go... But I doubt it.
You'll find few people who've done significant development with multiple generations of languages who don't prefer the later ones. This is because it makes *them* more productive. Google writing a spreadsheet as a web app might absolutely murder the hardware and consume resources like a madman, but I'm betting there was a lot less developer effort put into that cross-platform app than say openoffice.org.
Of course there will always be the speed freaks who prefer to develop at lower levels of abstraction and that's fine, they have a different goal to the majority of developers but I'm sure glad they're the guys who write the kernel I use daily, where spending an extra 40 hours development time for slight performance advantages actually makes sense.
I'm not sure how bolding makes your point, I didn't say it was a coke vs. pepsi thing I was saying it was a grandpa vs grandson thing.
It uses 400% more resources, big whoop. 9 iterations of moore's law and that amount of resources is just as negligible as the office client is now, let alone *manageable* which is the important number and where 100mb squarely sits at the moment. The *difference* the "magic sauce" is specifically in the automatic collaboration and portability enabled by the server-side nature. Whether it will be revolutionary or not time will tell but claiming it's just the same as before seems to me like sticking your fingers in your ears.
Wolfram alpha is kind of interesting, I've yet to coax it into answering a question I actually wanted the answer to though, traditional search engines do that for me every day though they do make it seem deceptively simple.
I'd still like to know what lesson we learnt from SMTP that wave hasn't attempted to address.
"instead of learning the lesson that SMTP taught US 25 years ago " which lesson in particular are they ignoring?
As for the thick app argument, why do you care? Seriously, if the solution works it works. The toggle switch guys scoffed at the punch card guys, the punch card guys scoffed at the interactive asm editor guys, the asm guys scoffed at the C guys, the C guys scoffed at the Java guys, the Java guys scoff at the Ruby/Python/PHP/JS guys. You don't see the trend?
You can gather what I'm talking about based on the context of what I was replying to.
In this case I was pointing out that if your server (X) allows for a transparently interchangeable IPC layer depending on your needs then most people would consider this to be *good* design and an example of a low degree of coupling, not high.
Keeping in mind that a display system is inherently client-server and it needs to have SOME model for IPC of course.
I agree that the installers don't have to have the painful behaviour that they do, I'm just pointing out that current operating systems make it easier for this to happen than they should.
Abstract the presentation of application launchers from the launcher entries themselves, and the installer maintainers have to go out of their way to be painful, rather than having to put in work to detect previous entries.
Professor Frink: Well, sure, the Frinkiac-7 looks impressive, don't touch it, but I predict that within 100 years, computers will be twice as powerful, 10,000 times larger, and so expensive that only the five richest kings of Europe will own them.
Summary:in terms of the entry cost to compete, and where big vendors like Amazon and Google are already divvying up the market, the odds of a small-fry, open-source start-up challenging 'Goliath' are slim. It's not a new argument: Nick Carr has been suggesting for some time that only a few, big companies can afford relevance in this hardware-intensive business
Sure, if you start off with an assumption that that statement is and always will be correct then there may be a problem. But uh, why are we assuming it's correct?
I mentioned Windows because he was talking about windows, you're right that all major operating systems handle desktop icons this way.
A more ideal solution would be to abstract the presentation of the desktop from application launchers. Much the way I believe Ubuntu handles the "applications" menu.
It's standard practice for any large data-warehouse to have a process of transforming their relational data into *something else* before it's used by whatever generates reports.
For me using an RDMS as the persistence layer for an object-oriented application has ALWAYS felt like a bit of a kludge. Like we're using it just because it's what we have, rather than the best tool for the job.
I look to the oil companies to innovate in drilling technologies. I look to financial companies to hopefully not innovate too much anywhere:-) I look to IT companies to innovate in IT.
I dunno about you, but I've seen an incredible amount of money spent in the last 10 years or so attempting to change those massive relational databases into formats that can be reported on, as well as huge amounts of energy put into moving from one relational schema to another.
Pretending the big conglomerates present the best answer just because they're big is a recipe for non-movement.
Imagine the company I work for spends 15 years debating a particular new growth strategy, where I had plenty of input and so did many other people and the eventual outcome was that majority/executive opinion decided on the course of action that I was debating against.
What would be the result if after this decision is made I continued spending my time formulating arguments against the decided strategy and disseminating them through the company as a whole? Swift kick up the arse would be the correct answer.
Indeed, most places in the world have had a list of banned books for a very long time. The nice thing about the Internet is it tends to make this censorship more obvious and people kick up a shitstorm about it. That doesn't always happen but it's better than it was.
HTML5 is a new scripting language? Or a plugin that does something "cool"?
That's user candy fella.
I pretty clearly was not describing a paradigm, I was simply pointing out that things developers like tend towards faster development leading towards faster roi.
Think about it. Who generally has the expertise and trust of management to make such decisions? If developers don't have the most input they certainly do have a say that holds influence.
'developer candy' can also be translated to 'lower barrier to entry' (cheaper programmers), 'faster ROI' (faster development for experienced programmers) and 'inherently higher quality' (larger cookie-cutter components)
What do you think developers enjoy working with? Inconsistent rickety unstable messes?
Fact is fella the end users don't decide on frameworks. And while sometimes decisions like that are made higher up often it is at the whim of a developer.
Slashdot Mirror
What the hell? Care to explain that little logical implication to me?
You have to wonder at the people who seem to think that every ATM is a hand-crafted flower for its specific context right down to the keypad too
Lack of inferiority doesn't necessarily imply superiority :-)
Personally I'm currently scrambling to go from a RoR project to a horrible C/BASIC/Canvas based ETL *thingy*. As always it's more about the project/how long you've been on it/what stage it's in than the specific technology.
Hey, maybe you're right. Maybe we've reached the endgame as far as computing abstractions go... But I doubt it.
You'll find few people who've done significant development with multiple generations of languages who don't prefer the later ones. This is because it makes *them* more productive. Google writing a spreadsheet as a web app might absolutely murder the hardware and consume resources like a madman, but I'm betting there was a lot less developer effort put into that cross-platform app than say openoffice.org.
Of course there will always be the speed freaks who prefer to develop at lower levels of abstraction and that's fine, they have a different goal to the majority of developers but I'm sure glad they're the guys who write the kernel I use daily, where spending an extra 40 hours development time for slight performance advantages actually makes sense.
Hmm. That got quite rambling.
I'm not sure how bolding makes your point, I didn't say it was a coke vs. pepsi thing I was saying it was a grandpa vs grandson thing.
It uses 400% more resources, big whoop. 9 iterations of moore's law and that amount of resources is just as negligible as the office client is now, let alone *manageable* which is the important number and where 100mb squarely sits at the moment.
The *difference* the "magic sauce" is specifically in the automatic collaboration and portability enabled by the server-side nature. Whether it will be revolutionary or not time will tell but claiming it's just the same as before seems to me like sticking your fingers in your ears.
Wolfram alpha is kind of interesting, I've yet to coax it into answering a question I actually wanted the answer to though, traditional search engines do that for me every day though they do make it seem deceptively simple.
I'd still like to know what lesson we learnt from SMTP that wave hasn't attempted to address.
"instead of learning the lesson that SMTP taught US 25 years ago " which lesson in particular are they ignoring?
As for the thick app argument, why do you care? Seriously, if the solution works it works. The toggle switch guys scoffed at the punch card guys, the punch card guys scoffed at the interactive asm editor guys, the asm guys scoffed at the C guys, the C guys scoffed at the Java guys, the Java guys scoff at the Ruby/Python/PHP/JS guys.
You don't see the trend?
You can gather what I'm talking about based on the context of what I was replying to.
In this case I was pointing out that if your server (X) allows for a transparently interchangeable IPC layer depending on your needs then most people would consider this to be *good* design and an example of a low degree of coupling, not high.
Keeping in mind that a display system is inherently client-server and it needs to have SOME model for IPC of course.
A layer of abstraction decoupling your system from the communication technology used with an overhead of 0 is poor design?
Shit, I need to re-read basically every software engineering book I own and re-evaluate career choices.
I agree that the installers don't have to have the painful behaviour that they do, I'm just pointing out that current operating systems make it easier for this to happen than they should.
Abstract the presentation of application launchers from the launcher entries themselves, and the installer maintainers have to go out of their way to be painful, rather than having to put in work to detect previous entries.
Professor Frink: Well, sure, the Frinkiac-7 looks impressive, don't touch it, but I predict that within 100 years, computers will be twice as powerful, 10,000 times larger, and so expensive that only the five richest kings of Europe will own them.
Summary:in terms of the entry cost to compete, and where big vendors like Amazon and Google are already divvying up the market, the odds of a small-fry, open-source start-up challenging 'Goliath' are slim. It's not a new argument: Nick Carr has been suggesting for some time that only a few, big companies can afford relevance in this hardware-intensive business
Sure, if you start off with an assumption that that statement is and always will be correct then there may be a problem. But uh, why are we assuming it's correct?
I mentioned Windows because he was talking about windows, you're right that all major operating systems handle desktop icons this way.
A more ideal solution would be to abstract the presentation of the desktop from application launchers. Much the way I believe Ubuntu handles the "applications" menu.
http://en.wikipedia.org/wiki/Star_schema
It's standard practice for any large data-warehouse to have a process of transforming their relational data into *something else* before it's used by whatever generates reports.
Correct. It's more of a problem with the "bare-metal" way windows handles start menu and desktop shortcuts for applications.
Saying RDMS's map object data well is a bit of a stretch, they map relational data well and that's it.
http://www.codinghorror.com/blog/archives/000621.html for some good background on the problems.
For me using an RDMS as the persistence layer for an object-oriented application has ALWAYS felt like a bit of a kludge. Like we're using it just because it's what we have, rather than the best tool for the job.
I look to the oil companies to innovate in drilling technologies. :-)
I look to financial companies to hopefully not innovate too much anywhere
I look to IT companies to innovate in IT.
I dunno about you, but I've seen an incredible amount of money spent in the last 10 years or so attempting to change those massive relational databases into formats that can be reported on, as well as huge amounts of energy put into moving from one relational schema to another.
Pretending the big conglomerates present the best answer just because they're big is a recipe for non-movement.
Imagine the company I work for spends 15 years debating a particular new growth strategy, where I had plenty of input and so did many other people and the eventual outcome was that majority/executive opinion decided on the course of action that I was debating against.
What would be the result if after this decision is made I continued spending my time formulating arguments against the decided strategy and disseminating them through the company as a whole? Swift kick up the arse would be the correct answer.
Indeed, most places in the world have had a list of banned books for a very long time. The nice thing about the Internet is it tends to make this censorship more obvious and people kick up a shitstorm about it. That doesn't always happen but it's better than it was.
HTML5 is a new scripting language? Or a plugin that does something "cool"?
That's user candy fella.
I pretty clearly was not describing a paradigm, I was simply pointing out that things developers like tend towards faster development leading towards faster roi.
I am a commercial software developer.
Think about it. Who generally has the expertise and trust of management to make such decisions? If developers don't have the most input they certainly do have a say that holds influence.
'developer candy' can also be translated to 'lower barrier to entry' (cheaper programmers), 'faster ROI' (faster development for experienced programmers) and 'inherently higher quality' (larger cookie-cutter components)
What do you think developers enjoy working with? Inconsistent rickety unstable messes?
Fact is fella the end users don't decide on frameworks. And while sometimes decisions like that are made higher up often it is at the whim of a developer.
samzenpus, you owe me a coffee and a new keyboard
There are at least as many representative democracies in the world that don't have a debt problem as that do.
Also just plain hilariously good at what they're *designed* to do.
Tow a trailer of wood with a fridge in the tray? No problem. But don't you be going and thinking you'll get that thing moving over 120km/h ;-)
dingdingding!
Democracy: the best we've got.
A pessimist is usually right but only an optimist will change the world.