3 ms·

Note that the 4 reasons you've listed have little to do with the technical merits of what you're working on. If anything, your point 2 and 3 talks more about th

by hiq 5y ago

Note that the 4 reasons you've listed have little to do with the technical merits of what you're working on. If anything, your point 2 and 3 talks more about the drawbacks of the technologies you're using rather than their advantages.

IanCal 5y ago
If by technical merits you mean "the tools all already exist" then sure, that's a drawback. But those tools don't create themselves, if nobody works in an area because they can't just install a settled on solution they'd never get built. ML didn't have the tooling it has now when I started with it, using recently released cuda to implement these new "deep neural networks" (more called RBMs at the time I think) on graphics cards because I needed them faster than the released MATLAB code - but now you can pip install huggingface and be away. When I was doing it, ML could have remained a niche area (people always asked me if AI was to do with aliens, and wondered what on earth I'd do with that degree) but has exploded after it hit critical thresholds. Maybe this will happen with crypto stuff. Maybe not. I've been very wrong on cryptos growth since the early days. But, people have business ideas and want them built. I'm building those while solving new interesting problems. Does that really sound so bad as a job? I'm not talking about my specific client work because I don't want anything to come across as shilling.
- hiq 5y ago
  > If by technical merits you mean "the tools all already exist" I mean the technologies that define this space, e.g. blockchains, ERC20 tokens, NFTs, and so on. I mean that in your comment, you didn't try to claim that these technologies were worth anything in themselves, just that there is a demand for them. > But, people have business ideas and want them built. I'm building those while solving new interesting problems. Does that really sound so bad as a job? No, but then you're also not in disagreement with people who find these technologies to be mostly a dead-end, since you just ignore how relevant these technologies are, you abstract away from the business requirements defined by somebody else.
  - IanCal 5y ago
    Because I'm answering about why I work in the space currently. You're reading into what I've not said. > I mean that in your comment, you didn't try to claim that these technologies were worth anything in themselves, What's the value "in itself" of postgres? The value comes from what you can build with it, what problems you can solve. > just that there is a demand for them. Because people want to build things and those are solutions to problems they face. That's why there's demand for them. > No, but then you're also not in disagreement with people who find these technologies to be mostly a dead-end, Again you're reading into things I've not written. Not knowing the long term future of this doesn't mean I agree it's mostly useless. > since you just ignore how relevant these technologies are Relevant in what way? To whom? They're certainly relevant if you want to integrate with any other projects in the area. >you abstract away from the business requirements defined by somebody else. The business requirements are defined by the business, yes. I can feel this building to a statement about how you can use other technology to solve any particular problem, so I'll come in front of that and say yes - you can always build things using other tools, blockchains aren't magic they're just code.

by hiq 5y ago

IanCal 5y ago

If by technical merits you mean "the tools all already exist" then sure, that's a drawback. But those tools don't create themselves, if nobody works in an area because they can't just install a settled on solution they'd never get built. ML didn't have the tooling it has now when I started with it, using recently released cuda to implement these new "deep neural networks" (more called RBMs at the time I think) on graphics cards because I needed them faster than the released MATLAB code - but now you can pip install huggingface and be away. When I was doing it, ML could have remained a niche area (people always asked me if AI was to do with aliens, and wondered what on earth I'd do with that degree) but has exploded after it hit critical thresholds. Maybe this will happen with crypto stuff. Maybe not. I've been very wrong on cryptos growth since the early days. But, people have business ideas and want them built. I'm building those while solving new interesting problems. Does that really sound so bad as a job? I'm not talking about my specific client work because I don't want anything to come across as shilling.

hiq 5y ago
> If by technical merits you mean "the tools all already exist" I mean the technologies that define this space, e.g. blockchains, ERC20 tokens, NFTs, and so on. I mean that in your comment, you didn't try to claim that these technologies were worth anything in themselves, just that there is a demand for them. > But, people have business ideas and want them built. I'm building those while solving new interesting problems. Does that really sound so bad as a job? No, but then you're also not in disagreement with people who find these technologies to be mostly a dead-end, since you just ignore how relevant these technologies are, you abstract away from the business requirements defined by somebody else.
- IanCal 5y ago
  Because I'm answering about why I work in the space currently. You're reading into what I've not said. > I mean that in your comment, you didn't try to claim that these technologies were worth anything in themselves, What's the value "in itself" of postgres? The value comes from what you can build with it, what problems you can solve. > just that there is a demand for them. Because people want to build things and those are solutions to problems they face. That's why there's demand for them. > No, but then you're also not in disagreement with people who find these technologies to be mostly a dead-end, Again you're reading into things I've not written. Not knowing the long term future of this doesn't mean I agree it's mostly useless. > since you just ignore how relevant these technologies are Relevant in what way? To whom? They're certainly relevant if you want to integrate with any other projects in the area. >you abstract away from the business requirements defined by somebody else. The business requirements are defined by the business, yes. I can feel this building to a statement about how you can use other technology to solve any particular problem, so I'll come in front of that and say yes - you can always build things using other tools, blockchains aren't magic they're just code.