Slashdot Mirror
Five PC Innovations the Industry Should Get To
An anonymous reader writes "Flexbeta.net has an article which describes 5 great technological advancements in computing that just about every PC user wants." From the article: "Why has there been such a sudden lack in innovation as of late? Are we in a technological drought? I like to stick to my own diagnosis of the industry as being too concerned with keeping a steady cash flow over social experimentation with new products but then again that's just an opinion from a little guy."
1. Better fans. Fast fans are going to make noise. There are quieter fans, and newer technologies like tip magnetic driven fans.
2. Better Cases. A BOTTLE OPENER?! What the hell? I stopped reading there.
3. Wireless everything. Sounds great until you realize wireless everything will probably conflict with your neighbor's wireless everything and the fact that encryption to keep your wireless everything will be another burden most users won't bother with. And of course, you still need power, so you're either back to wires or you have a lot of batteries.
4. More USB storage key uses. Already on the way via some new portable application standard. And, no, game keys won't work because you can still copy the files to other USB keys and thus the game's copy protection is worthless. They want you have to the actual CD (with their patented copy protection) because it makes piracy more difficult.
5. Store re-haul. Your hard drive is the same physical size because you probably want a lot of capacity that's really fast. If you could be happy with 5 gigs of storage that's pretty slow, you could have a smaller drive. And, yes, they're working on bootable flash drives.
I can't believe this is on Slashdot.
Servers that survive /.ing...
You keep using that word. I do not think it means what you think it means.
Patents, and Lawyers.
-----BEGIN PGP SIGNATURE-----
12345
-----END PGP SIGNATURE-----
Okay, some things like the USB key to function as a verifier (to avoid needing to plug a disc in for games) are a good idea, but I really think that he's asking a bit too much, too fast. I'm not fully versed in the development of today's hardware, but I do know for a fact that miniaturization costs money. That's the big reason why laptops still cost much more than desktops. Additionally, the wireless data transfer standards are still not sufficiently fast to support purely wireless connections. Sure, there are certain examples, but these are specific (like building 802.11b/g cards into printers?), but in general, stuff like Bluetooth can't handle the kind of speeds that consumers demand these days. And wireless monitors for near-consumer prices? Forget it!
I don't fault this guy for dreaming -- that's the stuff innovation is made of -- but I do fault him for thinking that companies seem to owe him this technology for some reason...
(Note: Slashdotted already?)
Someone will soon offer an operating system for free. Not Linux, but something like the Mac. Most likely, Google is going to release their own operating system. It won't have more features than Microsoft Windows. It will however, be more stable, and similiar to the Mac which is based on a UNIX core.
Since it is free, Google won't need to protect a monopoly unlike some other companies. That will encourage further innovation
I've used Windows, Solaris, and Linux. But if google made an os, I would switch to that pretty quickly.
"Why has there been such a sudden lack in innovation as of late? Are we in a technological drought?"
They are 100% right. I have a new dual core processor, with two 7800 GTX's running in SLI, 4 10,000 RPM Raptor Hard Drives in a RAID array running Windows x64 edition.
But the the real innovation these days is in quieter fans.
The key with innovation is that it usually doesn't come directly from companies, but rather academic and research based groups. Large companies merely buy and build upon interesting research work in order to create large profits.
Voice your opinion!
1. permanent read/write random-access storage that doesn't spin
2. ubiquitous ten-megabit wireless networking coast to coast
3. direct computer to brain link
4. batteries with 10 times existing capacity, or fuel cell that runs on common cheap organic liquid such as wood alchohol.
5. common-sense AI knowledgebase/engine to check spreadsheets, documents, databases for obvious errors.
I want faster load ups. I want a machine that turns on and boots instantaneously. I want games to start running the moment I double-click on them. I want my 2 GHz chip with its generation of software to perform quicker than my 400 MHz chip did with its generation of software.
Ok, I understand we can't all get what we want, so I want to know why what I want isn't happening.
Hey all, I'm the writer of the article and I just want to make a few things clear:
This list is just mainly things I personally have gripes with in the industry, not so much a "What's most important to do in the next 5 years" article.
I agree with you guys on the fact that there are many leaps and advancements in a lot of the technology sectors but I must say that in many ways, innovation and new ideas are not coming out like they used to.
It's great that they are building on the present technology but how many years do we expect them to re-tool the "same" thing over and over again until we demand something better and completely new?
Call my article bad or the "worst article ever" but again, this is just a playful list of things I personally would like to happen in the next 5 years and I would of included at least 10 more things but I'm a lazy bastard and I wrote the thing at nearly 3 A.M. before passing out on my desk.
Just...take it [the article] for what it is and try to honestly and truthfully discuss your ideas and wants for the future, because if no one talks about this sort of thing then things will just keep looking the same for the next decade without any real considerable change.
I don't think just about everybody wants those things. I think, for the most part, "just about everybody" don't realize options in computers.
The average user doesn't realize that fans can be quieter, or that a computer even needs to run cooler.
The average user just says, "That's the computer." when looking at the case. They don't think of a way that it could be better in any way. Sure, slashdotters do, but "just about everybody" isn't one of us.
The average user doesn't know how to plug things in (I just tell people plug things into the hole that it fits in and then plug the speaker in the hole with a picture of a speaker next to it and then they get it on their own), but they don't think of wireless everything. I tell (middle age adult) coworkers that my computer has a wireless mouse and keyboard and they're very impressed. They don't think of extrapolating like that.
The average user doesn't know what a USB key is.
The average user doesn't know much about hardware inside the computer at all (my website, ChristianNerds.com has a question and answer section, where they email me questions about computers and I try to answer them, and I get at least one person every week asking what a printer is... A Printer! "Um, it's the thing that prints out your stuff onto paper."). The average user doesn't know enough to know what else to want. They like faster and they like flashier graphics. That's about it. Oh, and music.
Luke
Quieter fans are out there. A couple years ago I kinda hit a wall with my case, I'd had the same case for like 6 years, and it just couldn't do the kind of cooling I needed without some fast fans. So I got a new case, figured while I was at it since the noise annoyed me I'd get some silent fans. I went from having 1 case fan to having 4 case fans, and the overall system noise dropped to less tahn half of what it was previously.
SilenX and Papst both make some excellently quiet fans that aren't too exspensive. They don't move quite as much air as some others, but still plenty to keep your shit cool if you properly plan airflow (good cooling comes from good airflow, not just sticking fans in randomly).
I can't read your article, so apparantly #6 should be "A server that doesn't suck" but from the mirrors of the first page I've seen and the /. comments, most of what you want is already out there.
Fans are easy, go buy quiet fans. There are plenty of companies out there perfectly willing to sell you quiet fans for your system. Mine is totally outfitted with them.
IF you want shit on your case, put it on then. Yuo can glue a bottle opener on, or add a tape deck, and so on. Some of your ideas aren't possible, like a soda dispenser (soda dispensers require large tanks of syurp, CO2, and a water hookup) but if you want your case to do more, make it do more.
Wireless everything, well go for it. You can get all your controllers wireless, and your speakers too. Your monitor, well sorry, but we don't have the technology to do 3+ gigabits over the air yet. PEopel keep working on faster wireless, but it's not at the level for monitors. Of course, even if ou do go wireless to the tower, you still have to have wires for power, or battries. You can't transmit enough power through the air to power a device efficiently, and physics is the problem there, no amount of innovation will solve it.
As for USB key uses, agan call it done. Many dongles for pro software are USB. Not done on consumer software because it would be more expensive than it's worth.
I don't know what your fifth was, and can't glean it from comments.
At any rate, it sounds like most of your personal beefs can either be solved now, or are things to which there are real, physical limits that prevent it from happening.
They do not invest in R&D (>1%) and competes on price. This forces everyone else to pretty much follow (except Apple) as there is only so much delta price people is willing to pay for innovations.
With Dell being the low-cost producer mainly due to Intel's discount this effectively means that the Innovations is being priced out of the industry.
Innovations is left to the componets suppliers and they pretty much only knows how to make things faster and smaller.
Help fight continental drift.
CLUSTERING
By now, you'd figure the Linux would might have gotten this down and perfected or at least out of the distro useable. We need mirroring failover, load-balancing, load-distributing, and task-distributing clustering all in one package. Some machines become on boot failover mirrors opertaing in synch with the others. Some machines on boot become drones for the first group balancing out loads without mirroring everything. Some more will become auxillary drones for overall load spreading to keep the core stable. And the last group will take various code to execute as needed by the first three layers.
THIN CLIENTS
There's no reason to stick insanely powered PCs in every corner of my house and inside every piece of audio-visual equipment, complicating heat disposal, electricity distribution, and network connectivity. Still all the guts in one place and put interfaces elsewhere. The Enterprise didn't run on thick clients with computers everywhere, it had a giant multiprocessing core and every lesser powered computational device around the ship was essentially an interface and some sensors and tools. We'll never see this future if we doggedly insist on sticking something comparable to a Cray of ten years ago in every little box. Our houses will go into electrical meltdown and our electric bills will become comparable to mortgages.
ENERGY EFFICIENCY
Sometimes around the end of the 1982 recession the world seems to have forgotten the lessons of the nonexistant phony baloney energy crisis: it is possible to do things that we need and like with less energy and without inane politically motivated changes in our lifestyle. It would be far better to have lower power processors and support chips, with multiple cores and each core hyperthreading on board and the chips working together if we needed the horsepower and the ability to turn those processors down when we weren't using them. We could also use lower powered graphics processors. We could use more efficient power supplies. The list goes on. With true hot plugging, we could in the OS software tell the mobo to turn down slots that had cards without any task at the time like a dial up modem only being used as an occasional fax. Tell the USB or Firewire drives to turn off until needed. As opposed to the current power saving systems that don't actually tend to work consistantly and without farking hard drives and the data stored there.
MODULARITY I don't call USB and Firewire everything modularity. I still have most everything jammed onto a single mobo and whatever isn't gets stuck in a PCI slot or one of the above mentioned busses. I would like to be able to power down, pull something like an Atari 2600 cartdrige out, and pop in another with a different processor. I'd like to be able to pop in more boards with no excess things I don't need. Like, say... a blade server. But it shouldn't cost fifty thousand dollars. We've had modularity, slots, etc. for a very long time now. Why is it that it costs insane amounts and is positioned in a way that discourages its use? Why must we be so monolithic? If my car was made that way, I'd not be able to stick a different air filter in without buying a new engine.
INTERFACE
How hard is it to understand that only somewhat accurate voice recognition, crappy voice synthesis, 3D and multi-monitor displays only for the well-heeled isn't cutting it? Instead we get convoluted eye candy keyboards, shiny mice and trackballs, we get geek candy. I want a speech recognition system that is speaker independent and given the Internet and sheer numbers of users, a wide range data base synthesizing the results together of millions of users should have by now come about. Nope. I was doing software based speech synthesis on a frigging 6502 with 64K RAM more than twenty years ago. Best we get is that voice of the MS Office assistants. Big deal we've had multi-monitor displays for years. No sign of them becoming the standard. So much
If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
but it seems that many here are mistaking improvement for innovation. Innovation is doing that which has not been done, nor was expected via the conduits of common sense.
... well, that is what I want. I'm not holding my breath.
Innovation in computing will take some doing. There are plenty of companies that are trying to find and accept new business models and methods, trying to adapt to new threats, both malicious and competition based, but there is no innovation per se'. Unless you want to count multicore processors, low voltage processors, battery and power technologies that are leaps and bounds above previous. These are arguably derivitive works, but they build cornerstones for true innovation.
Innovation in computing, by definition, must change how we use them in some way. The spreadsheet was an innovation. The DOS was an innovation. GUI OS was an innovation. What do we need now?
We need more human like interaction with computers. Grandma doesn't need to know what icon to click if the computer asks her what she wants to do? Little sheila doesn't have to know the innards of Google if all she has to do is ask what is the three main properties of an isotope?
There is an entire new (as yet unexplored) world of computing that is a huge layer between the user and the actual workings of a computer. All the recent 'innovations' in computing and technology bring us that little bit closer to the world of Star Trek computers. The people that help bridge what we have today over to what Star Trek and other futuristic folks have promised are the people that will bring innovation.
The computer is a tool. We use it in different ways, but it is a tool. It really doesn't matter what OS you use, it is still a tool. I envision robots interacting with humans, and in the background use the computer/Internet to help or assist humans. How many times have you asked somebody who that movie actor was? or what is the word that means so and so? or asked people around you what is a word that means blah blah blah? We are born, and grow up, and by accident of birth, we learn and experience what it is that makes us much of what we are (so psychiatrists say) but with the computer and Internet, that can change. When you can ask your robot or PDA what is the identity of the bird that I just heard, then you have innovation. When you can be shopping and ask your robot or pda if this camelback couch is a good deal or not, that is innovation.
When you can type out a shopping list for the grocery store, and a kid shows up when you get home with the items you wanted... that is innovation.
The point is that technology isn't innovation. Innovation is how we use the technology. You can surf progressively faster and faster, but if you continue to surf the same way, there is no innovation.
Perhaps some will argue with me (and TFA is unavailable) but innovation is not new batteries or a different design of laptop. Innovation is how we use the technology and information (that wants to be free by the way).
Innovation is how software makes the information more useful. Right now we still pay lawyers to do patent searches... computers should tell us if there is prior art or patents without paying a lawyer. Information is just information. Sure there are those that want you to pay for it, but any free information should be available in ways that is just not possible now... that is innovation. When your child can ask the computer how many bones are in the human hand, and be shown a picture of them on the 'face' of their personal robot, that is innovation.
Information doesn't want to just be free, it wants to be freely integrated into all of our lives. When there is even just one place in a rural 3rd world country where information like this is available, it doesn't take much to imagine that even the uneducated can ask for help finding a new way to solve a problem and finding how it was solved in all of history in other places. Say a third world company wants to build cars... and they ask the computer for cross reference of their design against all of the worlds minimum requirements for safety? If they got the answer, that is innovation.
Support NYCountryLawyer RIAA vs People
Have to agree, sadly.
I'm big into language and OS design research, but a new OS becoming a success is a pipe-dream. Rob Pike has a good essay on this.
Briefly, a new OS will never happen because as you said the barrier to entry is massive. I slightly disagree with where to place blame, though. It's not so much the fault of Microsoft/Linux/Apple dominating the scene as it is apathy and laziness in users.
Hypothetically, if OS/2 crushed Windows we would all be split between OS/2, OS X, and the Unix variety. Once an OS gets a decent amount of apps written for it, it's nearly impossible to leave it. This is also the cause for the backwards compatibility insanity of today. I've written an x86 assembler and was dumbfounded by the crap one must go through just to do what appears to be a simple task (translating mnemonics to binary). ModR/M+SIB is hell and deserves the Ugliest Hack of the Universe award. Microsoft, Intel, Apple, Linux, etc. They are all bound to backwards compatibility. That alone puts a big block on innovation. I'd love to just toss out the traditional WIMP interface some time. It will likely never happen.
I must say, I'm more than a little concerned about languages as well. There have always been many languages around, but not like today. COBOL, Fortran, C, Lisp, BASIC and a few others with small userbases were around 15-20 years ago. Today it is looking more and more like the Tower of Babel.
I'm tempted to say that the language design insanity (Perl, Python, Ruby, PHP, Java, JavaScript, etc. etc.) and the excessive search for the silver bullet is a response to the stagnation of software innovation. Why I say this, is because it is completely misplaced.
Perl is a response to a platform (Unix) that has hideous structured information. Or, rather, it has no structured information. Everything is unconnected text that sits in files. The closest thing to the ideal Perl replacement today would probably be SQL and a RDBMS. If properly integrated into the OS, I wager that 99% of the problems Perl solves would disappear. Hence, Perl would be unnecessary.
TCL, Python and Ruby (with Perl overlap)? Those are there because people tire of using low-level languages. C *is* Linux and Linux (Unix) is C. And of course, much of it politics which completely kills innovation. Python/Ruby aren't that far removed from Lisp, yet we must reinvent for political and egotistical reasons.
Look at Java for the worst example of NIH ever, and I don't see how any of it solves a single practical problem. The portability problem will never be solved by a VM, because everyone knows that the quality of the VM implementation is the key factor and varies widely. There is no difference between porting a C compiler to 100 platforms or porting a JavaVM to 100 platforms. The portability benefits are equal. Except Java has a seperate layer of indirection which adds further complexity and is a source of inefficiency. Sure, you can't distribute C apps like Java. But Java definately isn't the pinnacle of distributed applications.
The biggest area of innovation will happen vertically as it is today. Things like Final Cut Pro, Photoshop, Firefox, etc. They will continue to go off on their own direction. But essential system-level innovation is dead, I'm afraid.
On a side note, Linux employs the purest form of democracy seen to date. If you want something, you stick it in the system. This unfortunately does not make a well-designed system. A single person or a small team design the best systems. There are simply too many cooks stirring the pot to make it unifed and coherent. What we end up with is a pile of hacks on top of hacks. Nothing really well thought-out. It wouldn't be such a big deal if Unix was a system designed for evolution. It's not.. it's a big static ball of bits. Linux (and many other things) seem to require a recompile on the simplest changes to the system. And rarely does anything adhere to the Unix design philosophies (everything is a file, simple programs that
Dijkstra Considered Dead
I've used lots of software in my time that used hardware dongles and the only word I can think of to describe it was EXTREMELY annoying.
This USB key idea is pretty much exactly the same as the old parallel port dongle idea people once used.
I'm sure he would realize what I meant the day he had to start looking for that USB key about the size of a quarter he left on his desk somewhere... or the day he had to juggle what devices he had plugged into his computer.. after all, not everyone has 10 USB ports available to them.
Having one company in charge of the whole thing sounds like a disaster too... lets add a whole new layer of licensing fees to the mix...
"Yeah, so that won't deter pirates. So what? Nothing else does either."
Well, bingo. You've summarized there the _whole_ problem with this whole anti-piracy idiocy: it inconveniences everyone _except_ the pirates. It inconveniences _only_ the honest paying customers.
Now I _am_ firmly against piracy, and I'm proud to say that I legally own a copy (well, a license in software lingo) for every single piece of software on my computer. If something could actually deter pirates, I'd be for it.
But that's the whole point: it doesn't. Not only you can always find a no-CD crack or a warezed version, in most cases it's available actually _before_ the game hits the stores. Even the few people who still are too clueless to google for a download, will get the no-CD crack from a friend who knows how to.
And in the meantime it's people like me, people who actually paid for the game, who get to put up with hassles like:
- being locked out of a game I've paid for, because the CD got scratched.
- having my game screwed up without even telling me why, because some broken retarded piece of copy-protection was buggy and thought a legit copy was pirated. (E.g., Gangsters. Before the patch, if you had more than one CD drive, or had the game CD anywhere but in "D:", the retarded copy protection would think you're a pirate and throw all your gang members in jail. Repeatedly. No, I'm not kidding. It's too retarded to make up.)
- having a game crash to desktop periodically without any explanation, and after a month or so the devs come and say something like "uh, it's supposed to crash if it detects <insert brand of CD copying software> on your machine." Which I didn't, but apparently the copy protection was retarded enough to think so anyway. (Plus, let's get for a moment into the whole issue of them deciding for me which software I'm allowed to run on _my_ machine. How about they piss off and mind their own business?)
- being locked out of playing a game I've paid for in, say, Wine, because it comes with a retarded copy protection that wants to be loaded as a Windows driver or such.
Etc.
So now you propose, what? That for the few hundreds of games I legally own (yeah, literally. So I don't have a life), I should also dig through a big box of code-wheels and other retarded gizmos to be allowed to play? I hope you'll have some understanding if I'm a lot less than thrilled by that idea.
I wish they just stopped this idiocy completely already. It has one single job to do: deter pirates. If it doesn't do that, why keep such an annoyance around?
A polar bear is a cartesian bear after a coordinate transform.
You're thinking too classically, like a C programmer. With a higher-level language, more design intent can be expressed directly by the programmer, and used by the compiler for safety and optimisation purposes.
Sure, you can't check all arrays access for bounds violations, but if you think about it with your human level of understanding, very few accesses can ever really be dangerous. The trick is to be able to convey that level of understanding to the compiler, so it can perform the necessary checks, and no more.
As for NULL pointers, there's really no need for them at all if you have a serious type system. Recursive data structures are trivially dealt with if you have concepts like disjunctive types and pattern matching.
In fact, as useful as they can be at lower levels, pointers generally are only useful as reassignable references to objects. There's no need to relate pointers and arrays, nor to provide arbitrary pointer arithmetic and the NULL concept, with a moderately powerful type system.
There is a reason that many languages make a point of saying they don't support pointers, even if they have a more limited variation of the concept, as with things like Java. The killer isn't the concept of indirection, or changing the target, it's the arithmetic, and assignment of arbitrary values. That leads to a world without proper type safety, and it leads to aliasing concerns that have horribly negative effects on optimisation.
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.