munawwar

Don't know why people hating on GIMP.. maybe it's not exact replacement for photoshop.. but I've used it for product photo touch-up work and it gets the job done. Which is great job for free

I have deprecated my apps on Apple Store. I cant even put a message saying I am going to remove my app in couple months to my current users without another app review.

Anyway, after reading the comments here, I am glad I am not supporting this mess.

I am more in favour of going back to installing apps from the web, reading app reviews (also from the web), bringing back app listing websites (remember nokia days?), running an antivirus and cutting any chance of another monopoly rising.

Alternative app stores won't break this monopoly. Android for example already have alternatives.. but how many know them and how many use them? The one that's shipped with the OS always wins. Besides, most of them take like 20% tax.. which is still hugely profitable.

I've used python, pywebview and html/css on a (super) simple application recently. pywebview will use system installed web browser, rather than ship entire chromium. So maybe check that out.

My list begins with Firefox, Ubuntu & linux. I want to see OS, desktops and browsers not being solely controlled by for-profit corps. In the same line, mobile OSes don't have a good alternative. Any suggestions?

Also remember they take $99 / year / dev account just to open up access to their app store.. so add that into their revenue..

With 3% payment gateways, paypal etc made/makes profitable business.

30% of revenue is like taxes, you can build infrastructure for a country.

30% is like paying govt taxes... with which govts builds roads, infrastructure, what not. None of the app stores (including many of the alt app stores) needs that much.

>Apple protects its users better than the other major players.

If "other major players" are the baseline, then nothing is going to improve. They all stink.

For example, think why apple made it super easy to approve auto renewing payments but made it so hard to unsubscribe (which is hidden deep in settings).. if that's not dark UX then I don't know what is? Lots of scammy apps make use of this. Do they get removed from the app store inspite of all their negative reviews and customer complaints? mostly not.. best case, takes months..

Having client cache a subset of data is very much needed for the most frequently accessed data.. but it seems the various clients will yet have to implement this? (it could have been the best feature in the list)

Previously I had to implement this myself -> https://github.com/Munawwar/redicache

What do you think? Simpler than most implementations without sacrificing performance yes?

I am the author and I am aware that the number of hook based global state libraries are like increasing per week. But it only means that all the authors combined are onto something.

The simplicity of these hook based libraries are much much more appealing than say redux, mobx or even context api. Minimal setup time, simple to use.. not having to worry about potentially future performance issues (as with context api.. for those not aware, there is a sentence in react docs saying context forces re-render whenever the content changes regardless of shouldComponentUpdate rules)

The TypeScript part is simple too. Don't have to create classes or annotations. Create a type for your store data structure and start hacking away.

What do you all think?

Use any templating library that gives "template inheritance" as feature. I use Nunjucks (from Mozilla)

It's taken from here - https://github.com/kilimchoi/engineering-blogs

The author of that repo did the hard work of listing out some great blogs. (Many added their own blogs into the list later)

Any feedback/criticism would be appreciated :)

Good call. New post here - https://news.ycombinator.com/item?id=12280711. (Sorry people, for the repost)

Heh. I've seen someone in a similar situation in a previous organization I worked for. And the problem there was with a slow system admin team. The sysadm team was large, however the truth was most power and responsibility was given to exactly 3 of them. Yes, the 3 were experts and they did great work. However the sysadm team was shared by like 4 product teams (and sometimes also does some work for customer support dept).

From a manager's perspective, the root of his frustration wasn't that easy to solve. To "scale" the team, it was necessary for the 3 guys to take lesser work and train few/more of other sysadms so that they can delegate their work. However, their work was so pressing, they hardly had time/chance to teach others. (Also individual personalities may be different...some people are great at their work, but not so great at teaching others..some don't want to give up some power..some potential successors may not be willing to take up their responsibility etc..all of this increases the problem).

My guess is that your IT dept is in a similar situation. Explain to your manager(s) that they need to clear the traffic jam/bottleneck. Tell them to find and check the IT dept's backlog so that they are convinced. Tell that this inefficiency is potentially losing the company money. Also if your views are backed by some of your colleagues it will add weight to your words. And be patient (complaining too much back-fires, so save it). If nothing works out and the situation is unbearable..then..we'll try to find a different job?

I'd say that speeding up learning times for image related "AI" technology are important as well. Think of it..my nephew (1.5 year old) sees a cat just two times and is able to identify cats, whereas these neural nets needs huge training sets + performant machines and gpus.