exelib

But for column level permissions views are still needed, right? This isn't the case with hasura. (Of course hasura's approach has own limitations).

> Anyone accessing the generated API endpoint for the chat table will see exactly the rows they should, without our needing custom imperative server-side coding.

That looks like a bad joke. It is imperative, but now in the database. I'm not sure that it is better. Hasura's approach is declarative.

From my point of view, postgres' security model is still not suitable for users. It is more complex, imperative, and therefore more error prone. Hasura's approach is still not perfect, but a combination of postgres and hasura functionality make a huge difference.

Nope, they're saying that products like hasura resolve the issue with the duplication of data. Or to put in another way, you don't need multiple stores like kafka with redundant data storage. Just use hasura.

It shows only connections count and CPU. No other metrics, unfortunately.

In short, they aren't really comparable. The thing is, you use postgres RLS for DB users (or give access to an app). But hasura ACL is for app users.

I'm not OP, but I choose the first two. What are your recommendations?

Aren't p2p apps expose IPs of peers?

More info?

Two things:

1. Observe a great developer. How he/she think? How he/she approaches the problems? I was lucky and met Winsley back in 2008. Despite I was working only ~1 year for the company, this changed my whole life.

2. Do everything and try hard to do it great.

Please stop polluting our planet! There is no one advantage of this technique, but only disadvantages:

- It transpiles. Your claim is misleading and incorrect.

- On your local machine it doesn't make any difference. It's not faster.

- It has huge impact on the visitors. It transpiles every single visit. Every. It consumes power, burn energy through the window.

- It's slow. It's incredible slow and with every module it gets slower. Try it on a low-end mobile. Not on your shiny new iphone.

BTW, if you want unbundled and fast, look at snowpack or like.

From my point of view, it's a brilliant idea with huge potential! Even it's not sustainable in sense of customers, it's very great to build a "migration ecosystem" to get new customers with every new migration path.

> Nobody ever complained about the non-nativeness.

Noticed the same thing. Only devs care about "nativeness".

> The JVM is annoying to start up because it takes forever

In the wrong hands, yes. I didn't touch java for 5 years, but the start time was under 1 second. Is that forever? Look now at Quarkus and GraalVM.

Yeah, and it's very distracting if you want to copy a part from the url.

It's not a Postgres, but has a some limited Postgres dialect: CockroachDB. It has a range-based master-slave replication with routing and therefore you can use it instead of master-master replicated solutions. Be aware of high latency of distributed databases.

Well, where you store your JWT on the client? LocalStorage? Then you have probably much broader surface for an attacker.

In case of XSS you lost anyway. But with HttpOnly-Cookies the attacker can't steal your token and do everything from everywhere with your token.

Just my point of view:

I disagree with Ralf.

Scrum is here NOT to solving ANY of YOURS problems. It's like a car - it's give you ability to drive, but how you drive is up to you. If you can't drive - you drive nowhere. If the street is broken - you can't drive, too.

What Scrum do, based on it's pillars (Scrum Theory, http://www.scrumguides.org/scrum-guide.html), is ability (and force you) to solve your own problems. Not more. OK, there are some preventive things and some sort of problems just don't happen with Scrum (if you can drive, of course), but it's not goal of Scrum.

And because, I think, there is no alibis at all. It's just disability to drive.

/Scrum Master && Developer, who fight against large organization every f* day for ability to drive/

I watch "Peppa Pig", "Hello Kitty" and so on in English with my daughter :D But also watch conference talks (like Google.io) in English. I understand most of written text (except of Harvard Business Review - authors select not common used words) and a little bit worse talks in videos.

But it's opposite if I want to speak or write.

> You say you can't learn from audio/video courses. I hope that's not true.

Audio/video courses are to passive. After 5 minutes I can't remember me any more on rules or how to pronounce. I learn best if I participate actively.

Wow, that's nice idea! Thank you!

You need for every tool a good internet connection, if you need to sync big files. rsync is pretty good for big files because of possible compression, awareness of hardlinks and partial transfer. On top you have PKI and encryption.

I use unison from command line to sync files between pc's, backup server and mounted devices. Internally, if I remember me right, it's uses rsync.

I think no. I'm working on a application on which working more than 16 teams for more than 11 years. I think it is a large application.

It's OK to don't like it. But how much you worked with Python? In any (not really, but nearly) language you have some syntax and some formatting. In Java or like you use tabs or spaces to align your lines anyway, right? I have same align for semantically same lines. Why just don't use this _natural_ alignment for syntax and throw brackets away? One more benefit: There are no discussions about how formatting brackets. Same line? With space before? Next line like in C? Next line with alignment?

Why not? If I don't like to eat fish I just don't eat fish.

Personally I think, there are tradeoffs. After many years of PHP, I'm happy to not code in PHP for last 4 years (except a cs project). But for example for a lot of money I would code in PHP again. Why not? I'm also reject non-compiled languages for backend in big projects.

Why is speed up no option? - Yeah, SSDs are great. It's save x00.000€ (where x >= 1) in company I work for. - But also bottleneck and waste profiling while starting help us a lot. - Decide also raid0, raid5 or another raid levels to gain speed up. - Depends on technology here things like hotswap, jrebel, osgi or others. - Another thing is used DB. Maybe a memory DB is an option.

Without speedup there not so many options. - As already mentioned, test driven developing or like to reduce needness to start up - As you said, multitasking. It's doesn't work for me, if I try to do more work. But it's work if I do some not directly related tasks like writing mails, contribute to wiki pages, grooming, reading articles or HN.

Good point! Never burn a bridge.

Go. If your employer don't care about your training and other good things, don't spent your time too. I think it can be better for other guys, if you tell your current boss why you leave the company.

Why downvote without comment?

JVM starts in less than second also for a little bit more as "hello world" java program. But different crappy path/annotations scanning and bytecode manipulating tools like hibernate take a long time.

Zero-downtime for web servers/services is very easy. In most cases. Even if the web server needs minutes to start.

It's a trickier, if you (must) have persistent connection from client to (app)server and have a state. And a lot trickier, if you will upgrade (big, xxx gigabytes in size) schema or whole relatioanal database.