4 ms·

Not interested in video, text or it didn't happen. I run transactional systems, concurrent oltp performance or it didn't happen. We have hugely diverse sets o

Not interested in video, text or it didn't happen. I run transactional systems, concurrent oltp performance or it didn't happen. We have hugely diverse sets of interrelated applications and systems build either against the primary data set, down stream datalake/bi/cubes systems. Integration and compatibility or it didn't happen. We run Cloud or Linux systems. Platform support or it didn't happen. Some of us run databases that are decades old.. it needs to have been happening for at least 5 years and/or be being shipped by a major vendor or it didn't happen. See also: must have 10 years experience in technologies that have existed for 5. New tech is hard, data is important and we're busy, in an industry with many many new shiny things. Neither nice nor fair, sorry.

didgetmaster 4y ago
When I worked at Novell in the early 90s, one of the engineers ran a demo of some "new shiny thing" called an Internet Browser (I think it was the original Mozilla one). Several of my co-workers made comments about it that reminded me of your post. The management of Novell basically ignored the whole new "Internet thing" until it put them out of business. I am not saying that my data management system is going to put all those other DB systems out of business, but it is interesting how so many people on these tech forums seem to be against anything but the status quo.
- TristanBall 4y ago
  Yeah, tbh, it sounded harsher than i intended, and you could actually plot a pretty successful path by life simply by doing the opposite of everything I say. But there's lots of new shiny. Yes, some of it turns out to be the new big thing. Lots more of it turns out to a neat feature that simply gets replicated / bought and integrated into current platforms. Lots more settled into specific niches or makets you'll almost never hear about on HN. The majority disappears, wrong or right, and we all have time and attention budgets. I'm not against, I'm just not interested until it can tick more boxes. Or unless it happens to tweak my interest for whatever personal quirk of mine - and in this instance all the information is stuck behind my least preferred delivery style so....? Even then it's often moot.. the database I'm using now is dictated by application compatibility more than anything.. there's already "better" for almost any measure of the word, _except_ that this is the db our vendor app supports and have stored procedures developed for, and porting isn't a priority. All that said, if and when you take over the db industry, you're welcome to hold me up as yet another grumpy old timer who couldn't keep up!
warrenm 4y ago
>We run Cloud or Linux systems. Platform support or it didn't happen. This 100x I find it very strange that the demo requires Windows
- didgetmaster 4y ago
  While the system is designed to run on any platform, I simply don't have the bandwidth to support every OS for every build at this stage of the startup. We got it running on one Linux distribution for a few builds, but until we get more resources I have to limit its scope for now. We are looking for innovators and early adopters at this stage (for those familiar with Geoffery Moore's book 'Crossing the Chasm') who are willing to evaluate the technology without concern that every OS, every language, every integration, and every feature isn't working perfectly yet. I thought HN would be a great place to find such people, but based on several of the comments, there are a lot of late adopters and even laggards hanging out here.
  - warrenm 4y ago
    >there are a lot of late adopters and even laggards hanging out here Databases, data stores, etc all have to work flawlessly for a long time You're not going to find many (if any) "early adopters" in that field
  - TristanBall 4y ago
    I suspect the data science folks are more likely to be interested in what you've done than dba's, at the stage you're at. They're more likely to be working with desktop tools, with their own copies of unstructured or semi-structured data and they're far more likely to be doing as-hoc or exploritory queries and manipulations, and they'll care far less about concurrent access.. which is absolutely critical to just about any environment who still think it's worthwhile to hire dba's. Faster queries for a single user on a desktop solves no problems I have - as nifty as some of the techniques you're starting to talk about are.

didgetmaster 4y ago

When I worked at Novell in the early 90s, one of the engineers ran a demo of some "new shiny thing" called an Internet Browser (I think it was the original Mozilla one). Several of my co-workers made comments about it that reminded me of your post. The management of Novell basically ignored the whole new "Internet thing" until it put them out of business. I am not saying that my data management system is going to put all those other DB systems out of business, but it is interesting how so many people on these tech forums seem to be against anything but the status quo.

TristanBall 4y ago
Yeah, tbh, it sounded harsher than i intended, and you could actually plot a pretty successful path by life simply by doing the opposite of everything I say. But there's lots of new shiny. Yes, some of it turns out to be the new big thing. Lots more of it turns out to a neat feature that simply gets replicated / bought and integrated into current platforms. Lots more settled into specific niches or makets you'll almost never hear about on HN. The majority disappears, wrong or right, and we all have time and attention budgets. I'm not against, I'm just not interested until it can tick more boxes. Or unless it happens to tweak my interest for whatever personal quirk of mine - and in this instance all the information is stuck behind my least preferred delivery style so....? Even then it's often moot.. the database I'm using now is dictated by application compatibility more than anything.. there's already "better" for almost any measure of the word, _except_ that this is the db our vendor app supports and have stored procedures developed for, and porting isn't a priority. All that said, if and when you take over the db industry, you're welcome to hold me up as yet another grumpy old timer who couldn't keep up!

warrenm 4y ago

>We run Cloud or Linux systems. Platform support or it didn't happen. This 100x I find it very strange that the demo requires Windows

didgetmaster 4y ago
While the system is designed to run on any platform, I simply don't have the bandwidth to support every OS for every build at this stage of the startup. We got it running on one Linux distribution for a few builds, but until we get more resources I have to limit its scope for now. We are looking for innovators and early adopters at this stage (for those familiar with Geoffery Moore's book 'Crossing the Chasm') who are willing to evaluate the technology without concern that every OS, every language, every integration, and every feature isn't working perfectly yet. I thought HN would be a great place to find such people, but based on several of the comments, there are a lot of late adopters and even laggards hanging out here.
- warrenm 4y ago
  >there are a lot of late adopters and even laggards hanging out here Databases, data stores, etc all have to work flawlessly for a long time You're not going to find many (if any) "early adopters" in that field
- TristanBall 4y ago
  I suspect the data science folks are more likely to be interested in what you've done than dba's, at the stage you're at. They're more likely to be working with desktop tools, with their own copies of unstructured or semi-structured data and they're far more likely to be doing as-hoc or exploritory queries and manipulations, and they'll care far less about concurrent access.. which is absolutely critical to just about any environment who still think it's worthwhile to hire dba's. Faster queries for a single user on a desktop solves no problems I have - as nifty as some of the techniques you're starting to talk about are.