Please create an account to participate in the Slashdot moderation system


Forgot your password?
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×

Comment Re:What about heat dissipation (Score 1) 61

It's not surprising. Hierarchic topologies for moving things around aren't novel, even in computing. The shape of your CPU coolers, the fat tree topology in the CM-5 computer, the topology of Internet links, the veins on leaves, yes, and the human circulatory system - all work in the same fashion.

Comment Re:how about conference with relevant languages (Score 1) 27

The same reasoning in civil engineering would translate to saying that you only need a shovel and a mason's trowel to build anything. Yes, you can, the "competitive features" like improved insulation or just fashionable architecture don't depend on how you put the parts in place. But more powerful tools allow you to do more things in limited time.

Having said that, building those tools is complicated enough so it takes too much time and isn't generally done, be it building languages, debuggers, or whatever tool you need, basically equally. You do correctly, if obliquely refer to a well-identified problem of insufficient tooling even in those languages that are generally considered sufficiently expressive to not require additional syntax or semantics. Interestingly, this is independent of whether you're working in a more specialized language that has the concepts in question embedded in its core primitives, or in a less specialized language that requires you to first build these higher concepts (of interest to you) out of its lower-level constructs. It would even appear that the problem is roughly of the same scope regardless of whether you're building a new language (in the traditional "lexical-syntactic" sense, not just in terms of new APIs) or not, with the "standard approach" you're mentioning involving using non-specific analysis and debugging tools that are all-too-often only of marginal benefit because they don't provide the views that a large system might require to be more easily comprehended or modified, for example, by a newcomer. Building improved tools unfortunately requires some kind of model regardless of whether the model is explicit in form of another language or merely implicit in the code of the tools and the patterns of use of some API you're building. But the designers of programming environments can't possibly anticipate all the domains you might want to use their environment for, so even if you decide not to use or develop another language for an application, unfortunately, not much changes, there are still tooling problems to be solved. You've merely shifted the burden from one kind of tools to another kind of tools, and one could successfully argue that accessible techniques for building better tools (to bring them into the realm of what Eric Raymond refers to as "casual programming" in TAoUP) are highly desirable for overall productivity.

Comment Re:More useful than current (Score 1) 27

there is nothing done in new languages that can't be done in others

Then why aren't we programming everything in assembly? (But I agree that the greatest value today may not lie in the development of new completely general languages but perhaps in finding ways to formulate ideas in specific application domains more concisely while keeping readability so that you could achieve more complex but beneficial things using less brain time, including doing the same (in a circular fashion) to the development of such languages in order to simplify this process it actually made sense to do in practice and didn't involve getting a PhD in computer science. The world indeed doesn't need another rehash of C++ or Lisp.)

Comment Re:More useful than current (Score 1) 27

It's about latency. If you're in the same room with someone and any communication takes seconds and you can pick someone's brain with regards to things you're interested in, and perhaps get some new ideas and immediately check whether someone else has already tried them, that could save a lot of time for some ideas to be incubated that might take otherwise more time "brain-offline" to develop. Yes, learning things that have already matured can often be done more easily online at your own pace. That's probably not what conferences should be about. Oh, and of course, there's also the food. ;)

Comment Re:Molecular computing (Score 2) 61

I love Stanislaw Lem's concept of "the last generation computer". It may have been tongue-in-cheek in the time he wrote Fiasco (when the much-hyped "fifth generation" was "the Next Big Thing") but the concept feels increasingly relevant these days.

"This was a computer of the 'last' generation--last, because no other could have greater calculating power. Limits were imposed by such properties of matter as Planck's constant and the speed of light. Greater calculating ability could be achieved only by the so-called imaginary computers, designed by theorists engaged in pure mathematics and not dependent on the real world. The constructors' dilemma arose from the necessity of satisfying mutually exclusive conditions to pack the most neurons into the smallest volume. The travel time of the signals could not be longer than the reaction time of the components; otherwise, the time taken by the signals would limit the speed of calculation. The newest relays responded in one-hundred-billionth of a second. They were the size of atoms, so that an actual computer had a diameter of barely three centimeters. A computer any larger would be slower. The Hermes' computer did indeed take up half the control room, but that was for its peripherals: decoders, hierarchic assemblers, and so-called hypothesis generators, which, with the linguistic modules, did not operate in real time. But decisions in critical situations, in extremis, were made by the lightning-swift core, which was no bigger than a pigeon's egg."

Comment Re:In other words, Moore's law will continue (Score 1) 61

Lots of progress in CPUs was driven by the desire to use more transistors to run sequential programs for old ISAs and computer architectures faster than the older hardware could. Unfortunately, the number of transistors appears to rise superlinearly with the performance of our sequential (or "observably-sequential" ?) state machines, which means that the diminished increase in performance can very well be deceiving because we're not using the components in an optimal way.

Comment Re:In other words, Moore's law will continue (Score 1) 61

Does the price include such things as packaging? If you get more transistors in a single package, and perhaps lower costs of routing (both in terms of design time, because of one extra dimension to play with, and in terms of manufacturing), the total expense per transistor probably still decreases even if the transistor (on its own) costs exactly the same.

Comment Re:So much for the singularity (Score 2) 61

Technically, "singularity" doesn't have as much to do with Moore's law as some people might claim, since - at least unless I misunderstood something - "singularity" implies some kind of vertical asymptote which Moore's law, being merely exponential, doesn't have. This means that Moore's law is not a sufficient condition for reaching "singularity". There would ALWAYS have to be some other kind of mechanism involved that could very well work even in absence of Moore's law, for example some kind of increased insight into how to make a large-scale machine that would be "more than a sum of its parts" and transgress the boundaries of human intellect. But ALL that Moore's law could do for us is to make the machine smaller. It's not even certain it's a necessary condition. It doesn't give us any insight as to how to build a machine that would both be smarter than us and could further improve on itself. Moore's law is no substitute for our limited knowledge.

Comment Re:Fascinating. All of TWO relevant languages (Score 1) 27

Now you made me wonder...which ones are those two? :) (I admit that I'm personally increasingly leaning towards general language implementation ideas and perhaps metalanguages that can be easily ported on top of different platforms so that one didn't have to get skills that are less-than-generally useful and could still easily change jobs while retaining the ability to create useful structures in languages that generally don't include them - without having to switch your entire platform, which is often not desirable, or having to press your employer for a platform switch, which is often unlikely to succeed.)

Comment Re:how about conference with relevant languages (Score 1) 27

niche and egghead languages aren't how the world at large does things.

The fallacy of appealing ad populum applies to programming languages as well. Not to mention the fact that lots of people who are not doing things "how the world at large does [it]" are not exactly going to be talking about it a lot if they find themselves in a competitive environment. Why give your competition ideas? I mean, the probability of your competition "getting it" is often very low but still non-zero.

Comment Re: Uncle Larry's yachts have all been patched (Score 1) 11

Maybe defect-free software is too high a bar but painting yourself into a corner with 70M LOCs hardly seems like the preferred alternative, considering the well-know correlations between size and absolute defect count. If the religion of complexity dies today, it won't be soon enough.

Slashdot Top Deals

The clash of ideas is the sound of freedom.