Slashdot Mirror


The Next Big IT Projects From the University Labs (infoworld.com)

snydeq writes: From unstructured data mining to visual microphones, academic labs are bringing future breakthrough possibilities to light, writes InfoWorld's Peter Wayner in his overview of nine university projects that could have lasting impact on IT. 'Open source programmers can usually build better code faster, often because they have bosses who pay them to build something that will pay off next quarter, not next century. Yet good computer science departments still manage to punch above — sometimes well above — their weight. While a good part of the research is devoted to arcane topics like the philosophical limits of computation, some of it can be tremendously useful for the world at large. What follows are nine projects currently under development at university labs that [could] have a broad impact on the world of computing.'

29 comments

  1. Slashdot is dead by Anonymous Coward · · Score: 0

    I see now that it has all been a charade. I kept trying to read this site because it's all I knew from the golden age of WWW, but I must now accept that it's truly dead. And don't even THINK of presenting me with some commie-hipster-loving-alternatives.

    1. Re:Slashdot is dead by Anonymous Coward · · Score: 0

      You just noticed this? Slashdot has been derptastic for years. It's a place where clueless people pretend to be smart, which can be pretty funny.

    2. Re:Slashdot is dead by Anonymous Coward · · Score: 0

      Be careful what you say, for the autumn breeze breathes new life into the leaves.

    3. Re:Slashdot is dead by __aaclcg7560 · · Score: 1

      Or global warming spikes the temperature upward for a day and the trees dump their pollen in ecstasy.

  2. Young kids with something to prove by Anonymous Coward · · Score: 0

    Once you get into your 30's, you start to have family responsibilities and want to be paid what you're worth.

    1. Re:Young kids with something to prove by Tablizer · · Score: 1

      Once you get into your 30's, you start to have family responsibilities and want to be paid what you're worth.

      Your "worth" is less if you have a life. Corporations value de-facto slaves. It's why corporations like H1B's: they are stuck here, away from their regular home and family, with nothing else to do but work their asses off.

      There's a reason corporations no longer value loyalty and shrank rewards for loyalty such as pensions and "experience" raises: they want you to leave after you get a family.

      You also know that their buzz-word filled toy is a pile of crap, and they don't want to hear the truth: they like yes-men. Short-term-thinking PHB's want hit-and-run trophies.

      Your best bet is often to go to work for companies that have already been burned by fads and cheesy outsourcers: they value wisdom.

  3. Know your strengths and weaknesses by Tablizer · · Score: 4, Insightful

    I don't believe universities are very good at delivering "finished" products. Those with their heads deep in the arcane theory and bits often forget or don't know how to think about how regular users interact with products. Good UI's and feature packaging is hard to get right.

    But, creating new algorithms or code libraries for those outside with a better "product sense" is certainly something universities can and have done well.

    1. Re:Know your strengths and weaknesses by Anonymous Coward · · Score: 0

      Just universities? Even big "popular" brand can deliver "finished" products. Their feature package? Worthless annoying gimmicks like shinny led's on the motherboard? fuck that kind of "features".

      A lot can't even be rested and assured their motherboard purchase of whatever brand working without any problems straight out of the box and wasting time with RMing. And when that doesn't happen most can't even last a year past their warranty(Graphics cards are common case).

      Not to mention the firm/software side of their "finished" products.

    2. Re:Know your strengths and weaknesses by Austerity+Empowers · · Score: 1

      That entire sentence should have been deleted. It is senseless and highly subject to point of view. In my opinion (doing this for a living, in several different corporations), the absolute worst code is proprietary corporate generated code: from bosses who want something that works next month and doesn't give a shit about next year. Even bad open source, by comparison, is frequently better.

    3. Re:Know your strengths and weaknesses by Tablizer · · Score: 2

      All such opinions are subjective, until the point we can directly examine customer neurons in action, or at least have objective surveys. If my opinion is to be deleted because it's subjective, as you appear to request, then so should yours, and there would be no opinions nor objective info in existence either way. Nor does Slashdot preclude anecdotal and subjective evidence. It's not an academic research publication. Your complaint is baseless.

      That being said, "corporate" software tends to have better UI's in my observation than most open-source projects and university projects.

      I believe this is because a corporation MUST make the software somewhat usable or they die. In OSS and research projects, you have more room to fail, in terms of popularity of project.

      Thus, corporate software generally ranges from C- to A in "usability", whereas the others can range from F to A. D's and F's are weeded out quickly in corporate land because the producers of such go bankrupt if they keep it up (or down).

      Blender, Gimp, and Open Office Base (OOB) are examples of some OSS products which have/had horrid UI's, in my opinion.

      Take OOB, for example. It took me forever to find the set-primary-key setting. I compared it to MS-Access, and although MS-Access's UI is messy, they gave at least 3 different ways to set (open) the primary-key setting tool. They probably did some user tests, and after learning users had a hard time finding the key setter, stuck it in multiple places. It's a messy work-around, but better than OOB's.

    4. Re:Know your strengths and weaknesses by LinuxIsGarbage · · Score: 0

      My favorite example is "The Gimp". Aside from a ridiculous name that's not suitable for a professional environment (like a lot of FOSS). They over estimate their target audience. They assume their target audience is a non-existent Professional Photoshop Refugee group and not home users.

      In the end Paint.NET is an easier to use, more performant piece of software for Windows. Though I still use my version of Paint Shop Pro 5 from 1998 cause it runs like mad on my i5.

    5. Re:Know your strengths and weaknesses by Tablizer · · Score: 1

      Paint.NET crashed my PC. I'm gun-shy to try it again.

      However, I do agree that Gimp is overkill for a good many uses. I often use MS-Paint instead because it's less key/mouse-strokes. But Paint lacks gamma level, contrast, brightness, blur/sharp, save-quality-levels, and selected (cursor-controlled) blur/sharp/bright/darken. Add these, and Paint would satisfy 95% of my graphics needs. (I'm not a graphic artist, but do need to clean up and tweak photos and logos on occasion for web publish.)

      I use to use Paint Shop Pro in the 90's. It was pretty good for simple stuff. Do they still have a free version?

    6. Re:Know your strengths and weaknesses by Anonymous Coward · · Score: 0

      Open Source also seems more prone to "Learn my new Non-standard short cuts" that only make sense to me.
      "Just use the Command Line"
      And "My way is better" even if everyone else hates it.

      Corporate Software needs to be supported, and that costs money.

    7. Re:Know your strengths and weaknesses by Austerity+Empowers · · Score: 1

      All such opinions are subjective, until the point we can directly examine customer neurons in action, or at least have objective surveys

      No, I am agreeing with you, thus sentence should have been deleted from TFS, even TFA: Open source programmers can usually build better code faster, often because they have bosses who pay them to build something that will pay off next quarter, not next century.

      Not because they are opinions, because it is senseless for the objective of the summary and article and distracts the reader with a highly objectionable and contentious statement that is not relevant to the primary value of the article: good university projects that are succeeding in spite of academia's focus on the upper levels of the ivory tower. If you are going to make a statement like that, you can't just leave it in there like a big turd, you have to defend it.

      Also I'm talking about code, I won't argue with you on UIs or user experience in general. open source does not do that well. And if you want a design-in, or build a platform around open source, you better be ready to pay money. OSS devs will fix bugs they recognize (i.e. they see as their problem, not as your problem) and if they need to re-architect and change the plumbing & API it will be done, so you better have someone on the line to keep your platform up to date. Academia may or may not do this for you, depending on the size of your project and how interesting it is to the staff.

    8. Re:Know your strengths and weaknesses by Tablizer · · Score: 1

      Okay, I misunderstood your context. No hard feelings.

      Note I didn't see where the article gave evidence that the listed projects were already successful. It seems more of a forecast, like a stock broker recommending what they feel will be growing stocks in the future.

      And, some of them are kits and/or API's that others could use to build polished products with. That's probably the best hope.

      And you are right that OSS doesn't necessarily mean "free" or "low cost". If you really want certain features or strong support, you will probably will have to pay for it. Otherwise, you are at the whims of other customers with money and/or other egos.

    9. Re:Know your strengths and weaknesses by Anonymous Coward · · Score: 0

      I thought you might be overstating the level of delusion but holy cow: high-end photo manipulation; note the word ‘high-end,’ this is in results that can be achieved with GIMP and workflow it supports; high-end is not mid-or low-end: touching up some holiday photos a couple times a year is not what GIMP is made for;

      Until GIMP supports more than 8bpc, it is not suited for anything except touching up holiday photos a couple times a year (and for /b/tards making new meme faces). Where on God's earth did they ever get an idea to the contrary? CinePaint (a fork) supports 16bpc and the usual professional formats for such (.exr, .tiff), among other things. The problem is, the CinePaint folks managed to make the core of GIMP better while somehow making the UI worse. Maybe it's only used by postprocessing savants who have memorized a Very Specific Way of doing Very Specific Things, but the way CP implements copy/paste and zoom/scroll reaches "screaming, hair-pulling rage"-levels of bizzare. Gimp, at least, gets ctrl-c (copy), ctrl-v (paste), scroll wheel (go up/down) and shift-scroll (go left/right) correct.

      There's no real win, sometimes. Also, don't try to edit large pictures in Gimp, no matter how much RAM you have, unless you have an insanely fast SSD too. Because it stores undo data to a swap file, and it INSISTS this file be written to physical disk. This was not a pleasant discovery.

    10. Re:Know your strengths and weaknesses by xxxJonBoyxxx · · Score: 1

      >> Paint Shop Pro in the 90's.

      Look for "paint shop pro 6.02 abandonware" - something like http://www.oldapps.com/Paint_S...

    11. Re:Know your strengths and weaknesses by tehcyder · · Score: 1

      I do agree that Gimp is overkill for a good many uses. I often use MS-Paint instead

      There's something to add to my collection of "things I never thought I'd read on slashdot".

      --
      To have a right to do a thing is not at all the same as to be right in doing it
    12. Re:Know your strengths and weaknesses by Tablizer · · Score: 1

      I should qualify that. I use MS-Paint for a large portion of many types of image fiddling, and then GIMP to add adjustments not available in Paint. I've picked up some Paint shortcuts and tricks over the years. One can make shadowed text with Paint, for example, even though it's not a "direct" feature. Sometimes I work on a higher res version, and then scale it down for web pages etc. so that edge artifacts don't show up.

    13. Re:Know your strengths and weaknesses by LinuxIsGarbage · · Score: 1

      The problem is, the CinePaint folks managed to make the core of GIMP better while somehow making the UI worse.

      The UI from GIMP can get worse?

      Another potential "Free" (as in Beer) Windows image editor for home use is PhotoFiltre

  4. Reads like a google translation. by dmomo · · Score: 3, Funny

    Did anyone else have trouble following the quoted text in this summary? I hesitantly clicked through, but couldn't bear to even make myself scroll through to see what the click-bait list items were. InfoWorld is like the Buzzfeed of tech, only without the proofreading.

    1. Re:Reads like a google translation. by Anonymous Coward · · Score: 0

      Hey! Buzzfeed is hilarious on purpose.

  5. RIP Slashdot by Anonymous Coward · · Score: 1

    "Open source programmers can usually build better code faster, often because their have bosses who pay them to build something that will pay off next quarter, not next century"

    Can someone fix the grammar and logical errors in this sentence?

    1. Re:RIP Slashdot by U2xhc2hkb3QgU3Vja3M · · Score: 4, Funny

      "Closed-source programmers usually have to build code faster, even though it's barely compiling, because their bosses pay them to build something that will go on sale next quarter even if it's not ready to ship and will require monthly patches for years after that."

    2. Re:RIP Slashdot by Tablizer · · Score: 1

      Its open-sorce grammer and spailing, whaddya expekt

    3. Re:RIP Slashdot by Anonymous Coward · · Score: 0

      It's a throw-away prototype. Burn it with the Holy Fire of Misconception!

    4. Re:RIP Slashdot by RabidReindeer · · Score: 1

      "Open source programmers can usually build better code faster, often because their have bosses who pay them to build something that will pay off next quarter, not next century"

      Seriously. WTF????

  6. their have bosses by xxxJonBoyxxx · · Score: 1

    >> Open source programmers can usually build better code faster, often because their have bosses who pay them to build something that will pay off next quarter, not next century

    Huh? Open source is the "paid" side of the coding world? Commercial products take centuries to release? Please show me the straw man you're trying to build.

    1. Re:their have bosses by nicoleb_x · · Score: 1

      Open source sure is magic sauce. I thought agile would help improve development but now I find out that we could have just changed the license.