3 ms·
Perhaps the problem is lack of industry experience? I have plenty of ideas. All I see are problems that need solving. There are so many things broken in web/IT
by zapperdapper 9y ago
Perhaps the problem is lack of industry experience?
I have plenty of ideas. All I see are problems that need solving. There are so many things broken in web/IT that they can't be listed here. I have a file with literally hundreds of problems that need fixing and that's just scratching the surface.
Example, I think someone already mentioned this earlier but have you ever browsed a bunch of corporate websites - I have seen so many websites that are so horribly broken that it's just crazy - there are literally MILLIONS of broken websites out there. You could start by looking into why that is (I've been thinking about it and believe there's numerous business ops just in that area alone)...
I recently wrote about how the first company I worked for ran their entire operation (which was quite complex) with a few dumb terminals and a few Z80s clocked at 4MHz, running something called TurboDOS:
https://coffeeandcode.neocities.org/remembering-turbo-dos.html https://coffeeandcode.neocities.org/remembering-turbo-dos.ht...
I once wrote an entire oil rig data acquisition system in TurboBasic in a weekend. You could use ON TIMER to called a routine every second or so, no assembler required. ON KEY and ON TIMER gave you event driven programming (ok primitive but effective). There were things like INPORTB and OUTPORTB (I think) that could write directly to ports for control and data acquisition.
So are we going back in terms of productivity? What can you do about that?
Are you still waiting for someone to come up with a viable alternative to JavaScript?
You could almost come up with something by looking at trends like the move to multi-machine, multi-processor, multi-core, multi-threaded and there are so many things that could be moved to that model for scale. Then there are the tools to help profile and debug those things. Example: Elasticsearch - Lucene was/is a great library for search/indexing, but the Elastic search guy (Shay Banon) saw the opportunity to scale that to multiple machines.
Scale drives a lot of innovation: look at innovative technologies like Redis for example.
Another example - SQL databases. Have you ever set up a MySQL replication cluster and kept it running 24/7? Sure there are excellent services like compose.io and the new kids on the block like Cockroach Labs, but highly scaled database with ease of set up and management is not really a solved problem. Just keeping the darn thing up and running is often a challenge.
Like I said, I could go on and on...
p.s. yes, people are right - a bit of competition is not necessarily a bad thing...