3 ms·

The internet, or dare I say it, Information Technology itself, has not yet come to proper terms with Time Context as an implicit aspect of data. Much informatio

The internet, or dare I say it, Information Technology itself, has not yet come to proper terms with Time Context as an implicit aspect of data. Much information is not static, but rather is a growing or changing development from year-to-year. Using today's generic tools to search for digital information relevant to a specific version or time period of anything is very ad-hoc, hit-and-miss. And we're only a mere few decades in. I think the time context of data is going to become increasingly important and valuable.

ghaff 2y ago
Having been involved in creating content and deciding whether it should be retired or not, it really isn't about the money. Product docs can be locked to versions and archived pretty well; most sensible people understand the docs for 1.0 may not apply to version 3.0. But there are a ton of things on a company website that don't really apply to the here and now and will be increasingly out of date. And, if you just leave everything accessible forever, it pollutes the current information. I get that some people prefer everything is just left online but it gets harder and harder for all the folks who just want info that's fairly current.
- gmueckl 2y ago
  You can explicitly move old content to an archive and put big banners on it that it doesn't reflect current information. MS does an OK job of making it obvious that you're looking at documentation for an older product, for example.
  - ghaff 2y ago
    As I wrote, product documentation is versioned anyway so that's fairly straightforward. But a ton of other stuff is variably still relevant and can get in the way of finding what's most relevant and accurate for today. Whenever I would go to update something older, it would invariably take me longer than I planned because even things that were not flat-out wrong often didn't reflect current thinking, terminology, usage, etc.
    - crtified 2y ago
      In addition to what's most relevant and accurate for today, past context is also important; knowledge accrues. In computing we are used to fast technological redundancy, surfing the cutting edge of technology, where the greatest commercial relevance is. But one of my hobbies is vintage or retro computing, and something everybody in this hobby realises sooner or later is how much of an ordeal it is to collate even 20-year-old internet information. And that's in a digital interest area! Now do the same for news or the detail of recent history. Pick an issue and follow it's thread through internet history, year-to-year. It should be easy, shouldn't it? The reality is, half the sources aren't even dated. If we were designing an internet future, the ability to step back through time, in terms of digital information, would be a huge boon for humanity. Because, as critically important as it may be!, today is still only today.
      - ghaff 2y ago
        I don't really disagree. The way search evolved, it never had much of an explicit temporal component--probably because it didn't matter much early on. But now we have a situation where companies legitimately don't want old results polluting newer (usually more relevant to most people) results.

ghaff 2y ago

Having been involved in creating content and deciding whether it should be retired or not, it really isn't about the money. Product docs can be locked to versions and archived pretty well; most sensible people understand the docs for 1.0 may not apply to version 3.0. But there are a ton of things on a company website that don't really apply to the here and now and will be increasingly out of date. And, if you just leave everything accessible forever, it pollutes the current information. I get that some people prefer everything is just left online but it gets harder and harder for all the folks who just want info that's fairly current.

gmueckl 2y ago
You can explicitly move old content to an archive and put big banners on it that it doesn't reflect current information. MS does an OK job of making it obvious that you're looking at documentation for an older product, for example.
- ghaff 2y ago
  As I wrote, product documentation is versioned anyway so that's fairly straightforward. But a ton of other stuff is variably still relevant and can get in the way of finding what's most relevant and accurate for today. Whenever I would go to update something older, it would invariably take me longer than I planned because even things that were not flat-out wrong often didn't reflect current thinking, terminology, usage, etc.
  - crtified 2y ago
    In addition to what's most relevant and accurate for today, past context is also important; knowledge accrues. In computing we are used to fast technological redundancy, surfing the cutting edge of technology, where the greatest commercial relevance is. But one of my hobbies is vintage or retro computing, and something everybody in this hobby realises sooner or later is how much of an ordeal it is to collate even 20-year-old internet information. And that's in a digital interest area! Now do the same for news or the detail of recent history. Pick an issue and follow it's thread through internet history, year-to-year. It should be easy, shouldn't it? The reality is, half the sources aren't even dated. If we were designing an internet future, the ability to step back through time, in terms of digital information, would be a huge boon for humanity. Because, as critically important as it may be!, today is still only today.
    - ghaff 2y ago
      I don't really disagree. The way search evolved, it never had much of an explicit temporal component--probably because it didn't matter much early on. But now we have a situation where companies legitimately don't want old results polluting newer (usually more relevant to most people) results.