Building Network Automation Solutions
6 week online course starting in September 2017

More Features, Improved Lock-In

Found an interesting article on High Scalability blog (another must-read web site) on how PostgreSQL improves locking behavior in high-volume transaction environment.

Needless to say, the feature is totally proprietaryrather unique and not available in most other database products. Improved locking behavior ⇒ improved lock-in.

Moral of the story: Stop yammering. Networking is no different from any other field of IT.

Update: Yep, I goofed up on the proprietary bit (it was one of those “I don’t think this word means what you think it means” gotchas). However, if you think open source product can't have proprietary features or you can’t get locked into an open-source product, I congratulate you on your rosy perspective. Reality smudged mine years ago.

11 comments:

  1. That's not true. 'FOR UPDATE SKIP LOCKED' has been available on Oracle for some time.

    ReplyDelete
    Replies
    1. Thank you. My "black & white" perspective hit me again... Fixed.

      Delete
  2. How is this 'proprietary'? Just read the code (BSD licensed...) and implement it on your own in your own RDBMS. No license fee needed.

    EIGRP is proprietary. EIGRP is vendor lock-in.

    ReplyDelete
    Replies
    1. You're right - and I fixed the "proprietary" part. However, "implement it on your own RDBMS" part shows we live in rather different realities.

      BTW, I did implement my own RDBMS once and got it into production, so I know a bit what I'm talking about. Did you?

      Delete
    2. Why would I do that since I can just use BSD-licensed, fee-free PostgreSQL? A waste of time unless you want to study RDBMSes on your own. But have my congratulations that you did it on your own, impressive.

      What I mean - since you have the code available, license-free, why don't you give that to your company programers and ask for similar algorithm in your product? Sorry but I don't understand the "different realities" part - why do you think it's impossible?

      Delete
  3. EIGRP was sort of opened a few years ago:
    http://www.cisco.com/c/en/us/products/collateral/ios-nx-os-software/enhanced-interior-gateway-routing-protocol-eigrp/qa_C67-726299.html

    ReplyDelete
    Replies
    1. Q. What is Cisco releasing?
      A. Cisco is releasing the basic EIGRP to the IETF as an Informational RFC. This includes all the information needed to implement EIGRP, and its associated features, including High Availability (HA).

      So you lack more advanced EIGRP functionalities. To little, to late.

      Delete
  4. Perhaps by studying the history of server operating systems over the last 30 years, and the transition from vertically integrated servers first with proprietary operating systems (IBM mainframes, VMS, etc), then with proprietary integrations of open source (Solaris, HP-UX, AIX, etc), and in the end to disaggregated software (Linux, Windows Server) we can see that it is far cheaper per-unit to do the software once than it is to replicate hundreds of millions of dollars per year of R&D at every hardware vendor...

    Will leave it as an exercise for the reader to think through the industry structure under which the customers benefit from the industry only doing the software once. Hint: a single vendor who maximizes marginal revenue (that's a college level economics textbook phrase) is not that structure.

    ReplyDelete
  5. Claiming an opensource product introduces lock-in and is proprietary is rather dumb, didn't expect to see such a thing here. Don't believe me - just browse through https://github.com/postgres/postgres

    ReplyDelete
    Replies
    1. I could say the same thing about "claiming you can't be locked into an open-source product". The proof is left as an exercise for the interested reader.

      Delete
    2. Actually, while you can lock yourself into any product, including OSS, with OSS you can always find and switch to alternatives, so the lock-in isn't really a problem. If you spend a 6 figure sum of money to deploy openstack and use it organization wide, you are effectively locked into it indeed, because switching to something else, would mean another huge expense. There is, however, no lock-in in terms of integrations, whom you choose to get support from etc. Compare that to Oracle for example, who try real hard to lock you into their own OS, hardware and application stack, by declining support for anything else and you'll see the difference between vendor lock-in and what I would call tech-oriented self lock-in.

      Delete

You don't have to log in to post a comment, but please do provide your real name/URL. Anonymous comments might get deleted.