6 ms·

Effective knowledge capture is a huge problem in a lot of fields right now.

• nathanaldensr 10y ago
  I disagree. The real problem is that no one wants to pay for it. The tools are certainly there to capture knowledge, but everyone's trying to cut costs so much in a desire to optimize for the present.
  • ianai 10y ago
    Let's not forget all the people that would love to be the person receiving said information. But again, not happening.
  • dpcx 10y ago
    That doesn't disagree with the GPs comment at all. Whether it's for budgetary reasons or technological ones, the effectiveness of knowledge capture is still a huge problem.
  • CognitiveLens 10y ago
    Sounds like you actually agree, and have identified an underlying problem. The technical ability to capture knowledge, particularly for craft trades like glassblowing, has existed for as long as humans have been organized into societies - it's usually a question of dedicating resources to support it. The apprenticeship model needs a renaissance.
    • losteric 10y ago
      We need to get away from the dogma that everyone must and will get a college degree. Apprenticeships are as good, if not better, for many fields... especially in the world of software.
      • andrewflnr 10y ago
        Eh, it's going to be hard to replace a proper algorithms or computability class with an apprenticeship. I'd like to see a combined approach, classes in between an apprenticeship-style arrangement.
        • losteric 10y ago
          I don't think all programmers need a proper algorithms class. Think about construction. There are manual laborers with no domain knowledge that can do basic grunt work for pennies on the dollar, some with domain expertise to guide the grunts, and people that coordinate the entire construction project. Then there's an entirely separate group of people actually working on architecture... designing the building, ensuring the structure can hold load, meets customer and legal requirements, and can be delivered on a schedule. At my company, a good SDE1 hire is someone that's at the architecture-level... they might not be able to design a skyscraper but they can design a house. Yet 90%+ of their time will be spent putting up walls and installing plumbing. That's ridiculous for everyone involved.
          • p_l 10y ago
            That's more a disaster of our tooling - that we have menial work still, even though it was regarded bad and avoidable in 1970s with C! (not to mention smalltalk, lisp, etc.).
          • andrewflnr 10y ago
            They might get the thing done, but they'll do a worse job than someone who also knows about algorithmic complexity, CPU architecture, and other things that benefit from classroom instruction. We're not installing plumbing, we're doing mental work, where context matters. And I can't recommend that anyone aim for less than excellence.
            • eonw 10y ago
              plumbers and electricians, during their apprenticeships spend a LOT of time in classroom environments. they are even required to spend X hours per year to keep their skills up as well. so i would say they two are in fact very similar. classroom mixed with real world use of that knowledge is better then just sitting in class the entire time with no chance to get real world usage, help and walk-throughs, until after the fact. in which case apprenticeship might be better then a degree.
  • joe_the_user 10y ago
    Curious, how would one capture the knowledge of a good glassblower? As far as I can tell, automation changed manual skills two ways - by capturing and automating simple actions, it decreased the skills needed for a number of jobs but by freeing up resources up resources, it allowed some people to develop skills to a far greater depth than previously for specialized purposes such as science or art. But capturing the skills of these remaining highly skilled individuals seems like a tough task especially since there's no incentive to build a machine to duplicate their activities - the total value of the product they isn't that high. On the other hand, if someone figures out a way to 3d print high quality glassware, the whole field would go out the window.
• Someone 10y ago
  What makes cases like these extra hard is that the only place we know that is kind of knowledge can be stored in is human brains. If you want a cobol programmer (aside: iOS autocorrects that to cool programmer), get a decent programmer, give him a language reference and half a year of time, and presto! There's your cobol programmer. After all, programming is programming. If you want a good glassblower, there's no reservoir of people with experience in similar stuff. Having said that, there is at least one school that educates instrument makers of all kinds; Kamerlingh Onnes started one in Leiden because he needed instrument makers in his cryogenics lab.