Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Postman Now Supports gRPC

blog.postman.com

Post reply on HN
Print view View original

91–100 of 216 posts

Re: Postman Now Supports gRPC

#91

Post by robofanatic »

tluyben2 wrote: post #16
adhoc_slime wrote: post #14

I have a meta question to ask, in threads like these where big products announce new features or big funding I always see commentors lamenting bloated software or it having too many features, not doing the one simple thing you want it to do, etc. To me it seems the only trajectory of developers building products is to add new features, its how we justify keeping ourselves around most of the time. So what my question…

I definitely rather would have had a second (and third and further,...) app. But I guess most people do not.

I always wonder we have usable version numbers for API's or packages but why not for UIs. It would be nice to have the ability to switch the app to a specific version in the past.

Re: Postman Now Supports gRPC

#92

Post by nexuist »

judge2020 wrote: post #77
avel wrote: post #68

As I have also been frustrated with the direction Postman has taken, I have a great alternative to suggest, if the only thing you want to do is fire requests on demand easily and do not need automated testing. The VS Code Rest Client extension https://marketplace.visualstudio.com/items?itemName=humao.re... is great for testing and debugging APIs. You create a "my_request.http" file that contains something like POST h…

Can it do substitution with some sort of application-based KV store? The main reason I used a rest client like Insomnia is so that I can reference secrets like {{ client_secret }} or {{ api_key }} and have a change to the value in the KV store reflect for every request.

Yes, it seems like it:

"We support two types of variables, one is Custom Variables which is defined by user and can be further divided into Environment Variables, File Variables and Request Variables, the other is System Variables which is a predefined set of variables out-of-box."

However, what I can't figure out is if it supports programmatic variable setting. I have an API where I need to retrieve a token from an endpoint before using it for every other endpoint, and I'm using Postman's test scripts to set that token as an environment variable before sending all the other requests so they can all use it. I don't see a way to replicate that with this extension.

Re: Postman Now Supports gRPC

#93

Post by binwiederhier »

jonfk wrote: post #79

To add another data point, I also moved on from Postman a while ago. Similarly to several commentors here I had found it to have become bloated, slow and buggy imho. In my particular use case, I found the urlencoding when using oauth1 to be pretty buggy. It would sometimes not work on certain headers it generated. My team ended up moving to Paw but that has been a bit problematic since it only supports osx, so some t…

TIL about the intellij http client. Thank you for mentioning that. It looks awesome.blike Postman, but built-in to the IDE I have open anyway.

https://www.jetbrains.com/help/idea/http-client-in-product-c...

https://www.jetbrains.com/help/idea/exploring-http-syntax.ht...

Re: Postman Now Supports gRPC

#94

Post by BowBun »

franga2000 wrote: post #35
adhoc_slime wrote: post #14

I have a meta question to ask, in threads like these where big products announce new features or big funding I always see commentors lamenting bloated software or it having too many features, not doing the one simple thing you want it to do, etc. To me it seems the only trajectory of developers building products is to add new features, its how we justify keeping ourselves around most of the time. So what my question…

I really like how GitLab allows you to disable the container registry, issue tracker, CI and every other module if you don't use/need it. IMO this should be the way to introduce new features where at all possible.

Same with Github, you can disable issues or projects or whatever other features they have at a repo level.

I agree that this and feature flags are a great way to toggle things, but unfortunately for apps like Postman you still feel the load in the download and load times for the app. Usually those fancier features are also more code/data which results in straight increases in memory/disk regardless of whether you use something or not.

In IDEs like Jetbrains and VisualStudio, a lot of features are built like plugins which are a separate download and installation and allow you to disable them. I wonder if apps could take that approach to to keep more people happy.

Re: Postman Now Supports gRPC

#95

Post by CSSer »

jonfk wrote: post #79

To add another data point, I also moved on from Postman a while ago. Similarly to several commentors here I had found it to have become bloated, slow and buggy imho. In my particular use case, I found the urlencoding when using oauth1 to be pretty buggy. It would sometimes not work on certain headers it generated. My team ended up moving to Paw but that has been a bit problematic since it only supports osx, so some t…

Thanks for sharing. Is the code for httpyac anywhere? I can only find the repo for the docs. I'm trying to check it out, but I encountered some errors when trying it add it as a global with yarn because I'm running a node version greater v14. The extension also gave me a deprecation warning about Buffer() usage.

Re: Postman Now Supports gRPC

#97

Post by nexuist »

adhoc_slime wrote: post #32
woofcat wrote: post #25

Earlier quoted context omitted.

Things can be feature complete, and they often are at many businesses. You need to build a solution to problem x. You do that and then the project is over. Outside of maintenance work it's a stable code base and you the developer move onto a new problem. Examples of this in my mind are things like uBlock Origin. I'm not getting bombarded with updates on that project. It does the job it sets out to do, and it does it…

This would be what I ideally think should happen, a product hits a happy place and is finished. It seems, though, that when working in a business that you'll have many users with many use cases, you'll be competing with other software companies, the temptation is there to add the one feature more to please the user or compete a little better. this becomes a very specific problem each group of developers / business mu…

This is because you hired developers on payroll and it would be a dumb strategic move to fire everyone once the product is "finished." You have to keep building things in order to justify having people on payroll (otherwise you're paying them to do nothing every day).

If you take VC funding you're in for another world of pain, which is that your company may be perfectly profitable but if it isn't growing at a rate your investors want you'll get kicked out and replaced by someone who will take "product growth" seriously. In other words, Postman needs more paying customers so it needs to become more of a "one size fits all" tool.

One might think that the solution is to simply work on another product after Postman was declared "done," but it seems like nobody seriously attempts this anymore because the SaaS model is so much more profitable.

Re: Postman Now Supports gRPC

#98

Post by nexuist »

lambda_dn wrote: post #96

Bloatware that is a heave ugly GUI wrapper over cli tools like curl. Learn bash scripting and you will have unlimited power and speed vs this crap.

Learn bash scripting and you will hate yourself every day for having to wade through terrible language decisions made >30yrs ago while having to do everything by yourself because nobody else at your company knows Bash while also having no opportunities for advancement because nobody understands what you do and why it's important.

Re: Postman Now Supports gRPC

#99

Post by cannonpalms »

adhoc_slime wrote: post #14

I have a meta question to ask, in threads like these where big products announce new features or big funding I always see commentors lamenting bloated software or it having too many features, not doing the one simple thing you want it to do, etc. To me it seems the only trajectory of developers building products is to add new features, its how we justify keeping ourselves around most of the time. So what my question…

Companies must show revenue improvements or they die. A vanilla REST client is pretty damn hard to monetize. That makes the Postman team's decisions perfectly rational. It's simply the way the world works.
Post reply on HN