annnnd
annnnd on Hacker News
> should include optimization of interactions with the load balancer, e.g. where to terminate https, etc.
I'm curious - is there a case where you want to terminate HTTPS on end device instead of (only) on load balancer?
A year is actually not that much for porting such a huge legacy system to a new platform. I imagine most of the work was making the interfaces from/to other platforms.
I think parent was answering the usage of XML in this use case, which is not appropriate. XML has many strengths (as you have outlined), but it has also been (mis/ab)used so many times that it gained bad reputation. What my GGP suggests is an example of such. There is nothing to gain from XML here that any proper DB (with schema) wouldn't offer, or in this case, protobuf.
Kafka logs however are solving a different problem. The mental model is different - they do not record state, but the whole history of transactions, which makes it trivial to change the schema if/when need arises. Saying that the schema should be thought in advance and shouldn't change is not realistic IMHO.
I agree with you, but that is probably something that belongs to a plugin.
Which is the same as with anything else really. You need to practice to be good at something.
Ok, allow me to me rephrase GP: will this make it possible (in the long run) to train tensor flow models on AMD acceleration cards in a supported, easy-to-use and performant way, comparable to what NVidia is offering?
The key word is "eventually". CV is making huge progress lately, but it is nowhere near human level yet. My guess would be that the first generation of self-driving cars will use Lidar because it means shorter time-to-market. Later manufacturers will be able to supplement Lidar more and more with cameras + CV. And maybe microphones, thermal cameras,...?
Sure - but I am guessing a few errors here and there would have much less impact than a single programming mistake. Not that I think this is the right way, but I do understand where they come from.
Retention! Always focus on retention. If people keep using your product then you have product market fit. If they keep leaving - you have an uphill marketing battle.
I find it much easier to estimate the time it will take if I take myself out of the picture. When estimating, I always try to think "how much time would it take for my colleagues to plan, execute, test, etc... something"? It is almost impossible to judge your own speed, but with experience you learn how to estimate speed of others quite accurately.
Of course, I also let product managers know that any change will affect the deadline, and almost never in a good way. I do this every time they have another "really simple idea".
> Or that the only dev that willing to work with it are contractors that want money.
Greedy b*stards! </sarcasm>
Of course Go players probably saw it coming. Also, Google knew that he needed to leave good impression, not just beat the best human player.
This. They are not making paper out of stone, they are mixing limestone in plastic. No doubt there are some benefits to it (less plastic is used - though not 80% less because limestone has higher density), but marketing this as "paper" is highly misleading.
This is the perfect answer! Nothing else matters in this situation. The only problem with the picture painted by OP is that he is reluctant to let go of control, which is exactly what he should have done a year ago, at least partially.
Nobody in the company should be irreplaceable, otherwise you have a single point of failure.
On a more practical side: don't work on your vacation (and I mean this in general - like ever) or it doesn't make sense to even go on one. If you must, you can try shorter breaks first so you get confident that the world is still turning once you get back. And do not answer phone calls during the breaks!
Thank you for the clarification!
Genuinly curious: isn't the whole point of cryptographic hash functions that they are slow by design, so that they can't be made faster?
Actually, your experience sounds great! In general, it is much easier to find people with narrow interests. While it might seem to you that you are not employable, this is just an illusion - there are many developers, but few with experience "from development to product design to finance". This makes you a great fit for virtually any smaller company which doesn't have a separate person for each role - startups, new businesses,...
TL;DR: You didn't fail. You learned.
As for burnout - been there. If it's any consolation, it will go away once you find something new that will keep you motivated.
I respectfully disagree. Of course you need to
really understand backpropagation for any advanced stuff, but it is easier to ignore it at the beginning. Take Keras container, copy some MNIST example from somewhere and tweak it. Then, when you have a general feel of how training NNs works, gradually learn about each concept - BP should of course be one of the first. By the time you get into math stuff it will probably make much more sense because you will understand how it applies to your case.
But I guess the approach depends on how you best learn, so there is no wrong answer. Just jump in!
This just means that it is broken by design and will probably never be fixed. Nice.
Does anyone know about similar AMD vulnerabilities?
> ... human size robot called Romeo ... could be used for household tasks.
This seems like a very interesting choice of name for a household robot... :-)
In the old days this technique was used to get information about number of people who found the article interesting enough to read the next page, all without JavaScript. This of course assumes that readers notice the "next page" link, where is where they seem to have failed (or maybe the expectations of readers have changed so much that the new breed of readers no longer notice such links - quite possible).
I had no idea they have this, but it would probably be an overkill for SO. An ordinary (nVidia) GPU which starts at $150 is roughly >10 times faster at NN problems than 4 core CPU.
Yes, of course. I seem to remember that their data is still in their own datacenter though, so they might not be comfortable with using Amazon for this.
Why $150k? Ordinary GPU (<$1k) would probably be equivalent to 56 CPU cores, if not faster.
I imagine the reason for using CPU was that they would throw it out anyway, so there was no need to make an ideal solution - and they already had those 56 CPU cores.
And for those who, like me, still didn't get it - it was a slogan for a soap and reads "(99 + 44/100)%", in other words: 99.44% pure. Thanks TeMPOraL!
It is a wide field. There are people building tools (obviously needing strong maths) and people applying tools to real-world problems. Don't get me wrong, you will benefit from learning math behind the concepts, but being a competent software developer is in itself a rare and valuable feature. Jump in, it's incredible what we can do. It's fun. :)
Maybe this post should be "Show HN"?
I am happy for you, hope this grows!
However, may I suggest you add to description on GitHub page some more info? Performance is only a part of equation and description only talks about that. Did you give any thought to reliability and robustness? How about security? It is written in C++ - was it fuzzed? Does it support TLS? Authentication?
These are much more important questions (to me) at this time (disclosure: using Mosquitto and happy with it, didn't hit bottlenecks yet). I have no doubt you will address them sooner or later, but it might be good to start working on some of them as soon as possible - or, if you already have this covered, to write about it.
Of course, I might not be your target customer, at least at this stage, in which case please disregard this advice. :)