I don't know Symbian native development but I work with developers who say it was pretty decent at the time.
However, in my experience J2ME on phones was a horrible developer experience. The implementations were terrible, incompatible, incomplete, hard to get decent documentation for, even harder to get developer support for, easily crashed some phones (hey Siemens) etc. Write once, work nowhere. Nokias Symbian phones were the most pleasant ones I got to develop J2ME apps for. Yeah, it was still crap.
Currently, the accelerator pedal usually has a continuous effect - press a little deeper and you get a little more oomph, press more deeper and you get a lot more oomph (unless you are too low on the revs). That is very easy for the brain to grok, predict and modulate.
I don't think that it is a good idea to make this a non-continuous effect. Press a bit and you get oomph up to a limit (that is non-intuitive unless you watch the speedometer all the time instead of watching the traffic). Press a little deeper and you get nothing, press even deeper and you get nothing. Press even more deeper and suddenly you get more oomph again. Not intuitive to modulate.
You're preaching to the choir in the latter point. I agree that what we call AI (as a research field) today is far remote from how a biological brain works and that our neurological science does not have a complete, working understanding of how it works.
However, I do not agree that a human brain is the better solver for many problems. The human brain is incredibly versatile. It can do a lot of things - although poorly to mediocre. Some of its behaviour is detrimental to effective use. For example, a car driver is probably a safe driver if not prone to stress, cognitive overload, road rage etc. Asking an autonomous vehicle to have this behaviour built in by design seems badly thought through, in my view.
Has to be equivalent to a human mind. No compromises.
If we didn't have humans driving today, I doubt that "equivalent to a human mind" would be an acceptable requirement if we were to invent road traffic. I would be scared shitless by the thought of humans piloting these missiles if I wasn't already desensitized to them. In fact I still am scared.
You probably don't want your car to be mad at you because you didn't let it flirt with that BMW at the red light, or spending a majority of its CPU cycles calculating cryptocurrency because it is bored from driving the same route back and forth every day, or having a fit of road rage after a close call with a stupid Volvo, or finding a new younger owner?
"definition of a minicomputer as a machine costing less than US$25,000 (equivalent to $161,000 in 2018), with an input-output device such as a teleprinter and at least four thousand words of memory, that is capable of running programs in a higher level language"
I would say that this nvidia thing qualifies (with some reservation about the teleprinter).
Under his "guidance", the Web has transitioned from a network where people participated in its development and had control over how they consumed it to one where they no longer participate, have no control, and have become passive consumers.
The "old" technology hasn't been removed. I would even argue that it is more easily accessible than ever.
But human laziness, quick Dopamine fixes and broken net security, all exploited by corporate greed, make *us* choose to transition away from that mode of participation. If you live in a relatively non-repressive country you should blame the consumers, not the producers or innovators.
With this approach, AI researchers are following a road to something completely different from human intelligence.
Which is perfectly fine. If you can disregard the debate around the term AI, this road can lead to (and already has) *better* solutions than humans can provide unassisted - for specific problems and quality definitions.
I don't believe that having one machine, one algorithm to solve all different problems is the right aim at all. And I definitely don't think that a human brain would have been the best target model if that was the aim.
Making a proper synthetic brain implementation is an interesting endeavour in itself and will provide a lot of useful understanding in medicine, sociology, technology, and a lot of other fields. As a solution to any concrete problems, I think that we can come up with a better solution than the brain for virtually any problem that we can formulate.
Desktop hardware usually last longer than the 0-4 years of updates that mobile OS:es get from their hardware vendors. We wouldn't want to throw away more kg:s of hardware every year just because we want security fixes or support for newer hard- and software.
... but then they blew it right back to 2019 on the very last row of the page:
"This film is not yet rated Filmratings.com MPAA Terms of Use Privacy Policy Your California Privacy Rights Children's Online Privacy Policy License Agreement Interest-Based Ads Marvel Insider Term"
Polyglocy also has its costs. Wrapping C++ code to be able to use it from Python is a non-zero cost in effort and maintenance, and so is mastery of several languages and frameworks. That cost should be taken into account and it may or may not sway the balance.
So very much this! Well evolved, tested, tried, fixed, documented, supported go-to libraries with a developed community and user base. The path to good sleep at night.
I too prefer that to having my application break because I used some random developer's left-padding component that I found in a repository.
Java is really not bad in this light. Remember, it is the tractor that pulls the race car out from the bushes when it has slid off the track.
Slashdot Mirror
They intended to *download* the contacts but actually uploaded them instead?
There are contenders to the title: https://www.telegraph.co.uk/te...
Yup: https://en.wikipedia.org/wiki/...
http://www.humoar.com/fake-roa...
Well, GIF has the benefit of having an intuitive pronunciation (well, two it seems) while PNG doesn't.
It is all connected. The "Jif" brand used to be called "Vim" in my country.
I don't know Symbian native development but I work with developers who say it was pretty decent at the time.
However, in my experience J2ME on phones was a horrible developer experience. The implementations were terrible, incompatible, incomplete, hard to get decent documentation for, even harder to get developer support for, easily crashed some phones (hey Siemens) etc. Write once, work nowhere. Nokias Symbian phones were the most pleasant ones I got to develop J2ME apps for. Yeah, it was still crap.
Currently, the accelerator pedal usually has a continuous effect - press a little deeper and you get a little more oomph, press more deeper and you get a lot more oomph (unless you are too low on the revs). That is very easy for the brain to grok, predict and modulate.
I don't think that it is a good idea to make this a non-continuous effect. Press a bit and you get oomph up to a limit (that is non-intuitive unless you watch the speedometer all the time instead of watching the traffic). Press a little deeper and you get nothing, press even deeper and you get nothing. Press even more deeper and suddenly you get more oomph again. Not intuitive to modulate.
You're preaching to the choir in the latter point. I agree that what we call AI (as a research field) today is far remote from how a biological brain works and that our neurological science does not have a complete, working understanding of how it works.
However, I do not agree that a human brain is the better solver for many problems. The human brain is incredibly versatile. It can do a lot of things - although poorly to mediocre. Some of its behaviour is detrimental to effective use. For example, a car driver is probably a safe driver if not prone to stress, cognitive overload, road rage etc. Asking an autonomous vehicle to have this behaviour built in by design seems badly thought through, in my view.
Has to be equivalent to a human mind. No compromises.
If we didn't have humans driving today, I doubt that "equivalent to a human mind" would be an acceptable requirement if we were to invent road traffic. I would be scared shitless by the thought of humans piloting these missiles if I wasn't already desensitized to them. In fact I still am scared.
You probably don't want your car to be mad at you because you didn't let it flirt with that BMW at the red light, or spending a majority of its CPU cycles calculating cryptocurrency because it is bored from driving the same route back and forth every day, or having a fit of road rage after a close call with a stupid Volvo, or finding a new younger owner?
How about some compromises?
He's now gearing up to plunk down more money on another pair...
Yeah, that will really teach Apple a lesson!
Oblig. https://xkcd.com/1205/
there's nothing wrong with doing marketing
Citation needed.
"definition of a minicomputer as a machine costing less than US$25,000 (equivalent to $161,000 in 2018), with an input-output device such as a teleprinter and at least four thousand words of memory, that is capable of running programs in a higher level language"
I would say that this nvidia thing qualifies (with some reservation about the teleprinter).
Under his "guidance", the Web has transitioned from a network where people participated in its development and had control over how they consumed it to one where they no longer participate, have no control, and have become passive consumers.
The "old" technology hasn't been removed. I would even argue that it is more easily accessible than ever.
But human laziness, quick Dopamine fixes and broken net security, all exploited by corporate greed, make *us* choose to transition away from that mode of participation. If you live in a relatively non-repressive country you should blame the consumers, not the producers or innovators.
With this approach, AI researchers are following a road to something completely different from human intelligence.
Which is perfectly fine. If you can disregard the debate around the term AI, this road can lead to (and already has) *better* solutions than humans can provide unassisted - for specific problems and quality definitions.
I don't believe that having one machine, one algorithm to solve all different problems is the right aim at all. And I definitely don't think that a human brain would have been the best target model if that was the aim.
Making a proper synthetic brain implementation is an interesting endeavour in itself and will provide a lot of useful understanding in medicine, sociology, technology, and a lot of other fields. As a solution to any concrete problems, I think that we can come up with a better solution than the brain for virtually any problem that we can formulate.
I am fascinated by how some people use circular definitions of ambiguous terms in order to argue for their personal belief:
"If we declare that one prerequisite for being an artist is that it is of the species Homo Sapiens, then something else cannot be an artist."
Kinda like the worn out one we are used to here on /.:
"If we define intelligence as something that only something made of meat can possess, then something not made of meat cannot possess it."
Desktop hardware usually last longer than the 0-4 years of updates that mobile OS:es get from their hardware vendors. We wouldn't want to throw away more kg:s of hardware every year just because we want security fixes or support for newer hard- and software.
You thinking of this? https://linux.slashdot.org/sto...
all of which are notorious for memory access bugs (in the runtime environment, in the case of Java).
Android does not run a JVM as far as I know, but Dalvik. And the only famous JVM memory access bug a five-second search gave me was from 2002.
... but then they blew it right back to 2019 on the very last row of the page:
"This film is not yet rated Filmratings.com MPAA Terms of Use Privacy Policy Your California Privacy Rights Children's Online Privacy Policy License Agreement Interest-Based Ads Marvel Insider Term"
If an app could expose this security hole then the OS is not secure by design (or implementation).
If it is not fixed at the OS level we just have to wait for the next similar vulnerability to be discovered.
I agree in general, though as usual "it depends".
Polyglocy also has its costs. Wrapping C++ code to be able to use it from Python is a non-zero cost in effort and maintenance, and so is mastery of several languages and frameworks. That cost should be taken into account and it may or may not sway the balance.
So very much this! Well evolved, tested, tried, fixed, documented, supported go-to libraries with a developed community and user base. The path to good sleep at night.
I too prefer that to having my application break because I used some random developer's left-padding component that I found in a repository.
Java is really not bad in this light. Remember, it is the tractor that pulls the race car out from the bushes when it has slid off the track.
I think that is showing the continued strong performance of Python taking over some roles Javascript has been taking over.
I would guess that the Python upswing can largely be attributed to it being very popular in machine learning.