Andy Bennett

@databasescaling

Data | Strategy | Delivery ⇾ Technologist & Engineer. Data-intensive systems. Founder . Inhabits the void between hardware, software and users.

London, UK
Vrijeme pridruživanja: ožujak 2010.

Tweetovi

Blokirali ste korisnika/cu @databasescaling

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @databasescaling

  1. Prikvačeni tweet
    13. pro 2019.

    troff is the standard typesetter.

    Poništi
  2. prije 9 sati

    I've now rearranged my diary several times in expectation of their call. This has been going on for weeks and weeks now.

    Prikaži ovu nit
    Poništi
  3. prije 9 sati

    Still having *major* problems with 's (lack of) customer service. I explain the problem, they offer a call, they don't call, they offer an explanation for a different problem, claim the case is settled and then we have to start again. ...which takes another week.

    Prikaži ovu nit
    Poništi
  4. prije 11 sati

    Yay! Handegg is 's word of the day. :-) #+

    Poništi
  5. 2. velj

    OH: "He is like a race horse fueled entirely by whisky."

    Poništi
  6. 29. sij
    Poništi
  7. 23. sij

    After much back and forth the message seems to be "we've can assure you we've done it but if you send us your bank statement we'll ask our team to look into it". So you haven't actually checked at your bank yet? I've checked with mine three times.

    Poništi
  8. 22. sij

    Despite assurances from I still have not received the promised refund after more than two weeks...

    Poništi
  9. proslijedio/la je Tweet
    21. sij
    Odgovor korisniku/ci

    rice-cookers are an unappreciated magical convenience based on simple magnets and they help make me more productive than slack ever has 😋

    Poništi
  10. 20. sij

    Solutions that fail to address all 4 will not revolutionise the marketplace. At best, they will provide another option among many that will be used to produce proprietary data models with proprietary tools.

    Prikaži ovu nit
    Poništi
  11. 20. sij

    As we've seen in other areas (e.g. instant messaging, events / calendaring), standards and interoperability are difficult and not enough. To succeed, users need a community (Data) and a reason (Incentives).

    Prikaži ovu nit
    Poništi
  12. 20. sij

    Which brings us on to 4) Incentives People need a reason to adopt the new thing. Perhaps it's better tooling for yourself. Perhaps it's better tooling for your team. Perhaps it can tell you more about how your data is being used by others. Perhaps it has lots of data you need.

    Prikaži ovu nit
    Poništi
  13. 20. sij

    So the system needs to work with the status quo. It be sympathetic to existing data management practices so that people can adopt it with their existing data.

    Prikaži ovu nit
    Poništi
  14. 20. sij

    3) Data We need examples! It's a Chicken and Egg problem. How can you have data if you don't have a way to publish it? ...and how can you make a good way to publish if you don't have any good examples?

    Prikaži ovu nit
    Poništi
  15. 20. sij

    So 1) Standards and 2) Interoperability ...are not enough. We need more. ...but what?

    Prikaži ovu nit
    Poništi
  16. 20. sij

    The current state of the art seems to be XML, JSON or CSV. These don't work at the ecosystem level tho'. Even within well defined niches, all the data publishers have their own profiles of the standard.

    Prikaži ovu nit
    Poništi
  17. 20. sij

    ...and even now the actual parts they interoperate on are fairly limited. How do we build something that allows generic data model interoperability at the ecosystem level?

    Prikaži ovu nit
    Poništi
  18. 20. sij

    For (very) simple stuff, things are pretty good. For things that are a very light model of a real world concept it's OK. Then we have things like Calendaring. Having an ecosystem of interoperable calendaring products took *decades*.

    Prikaži ovu nit
    Poništi
  19. 20. sij

    Finally we have Semantic Interoperability. Do we actually agree what things mean? Do we agree how those things map to the real world? Do we know which things are "the same" and which are not? Do others agree with us on that? This is where things get really sticky.

    Prikaži ovu nit
    Poništi
  20. 20. sij

    Are we extensible? i.e. can we agree representations for new things without changing the standard? Technology in this area tends to be OK within its niche but there's a long way to go to make it more user friendly.

    Prikaži ovu nit
    Poništi
  21. 20. sij

    Next we have Representational Interoperability. Do we both have the same understanding of data types? Do we both know what a number is and how it is represented? What about strings of text?

    Prikaži ovu nit
    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·