Postman acquires Akita for automated API observability
blog.postman.com
blog.postman.com
The collaboration features aren't useful on any size team.
I find this question really annoying. Almost all tech companies have no moat except for maybe brand and network moat which postman also kinda has. What's Apple's moat? Or Google's?
Also for Apple while I am not sure, but looking at the Android phones there is hardly any feature of Apple which hasn't been covered by some Android phone, so I don't think IP is an issue. It's just that overall quality is much more consistent for Apple across generations that it has built a good brand.
Every company also has a valuation, either through public markets, private markets, or discounted cash flows.
A house cleaning company has a moat of trained cleaners, existing client generation process, goodwill of existing clients, etc. If the value of their discounted cash flows exceed the cost of crossing their moat, they are vulnerable.
The argument is that crossing postman’s moat would cost much less than their private valuation, and their future roadmap is unlikely to build a moat that is significantly harder to cross.
The definition of a moat.
Same with postman. Right now they are so much ahead in features that it would take multiple years for other company to match it.
I agree 100%. I just said the same is true for postman that it is hard for some other company to replace it.
Google picked "How do we save (an additional) 20B+ per year in cash payments to a competitor" and "how do we have a path to accessing the world's data ... that is on a phone." -- and then won at both.
Apple has a strong luxury brand, higher quality, a large valuable market with a strong app community. They have a similar insulation to external players. Macbooks also tend to be popular in development communities due to its unix based os and strong support, which can be a strong alternative to a linux os for companies, but has been watered down with Windows push to support WSL2.
Postman doesn't have a moat. They don't have a strong marketplace where they can profit of exchanges. They don't have a significantly more advanced tool. it can be easily replaced by any user with an equivalent product with no real loss.
And they're competing with free and good enough but they're making a suite out of the tools and slapping a GUI on it. They're owning making it all work similarly and owning the training for their tools and if you pick them for all they do, it will be consistent.
How many companies use Atlassian suite even though it really just bundles a bunch of functionality that was already free and good enough? (ROT13 spoiler: nyzbfg nyy bs gurz!)
For what I need, it's much better and finally I can use multiple certificates for the same https domain for mTLS, which was a huge limitation in Postman
https://www.jetbrains.com/help/idea/http-client-in-product-c...
It might, but I think the appeal of Postman is how straightforward it is to play with APIs across basically anything in the HTTP stack, with a nice GUI. Also, I thought Postman also had the ability to export your queries to curl if you want to need to use it outside Postman.
Pretty much applies to a lot of startups. VC money does weird things.
I prefer this VS Code plugin for the very simple stuff https://marketplace.visualstudio.com/items?itemName=humao.re...
I'm sure there's a gap for a "better postman" that's more complex that that, but a lot less encrusted than postman.
However IMHO, the more complex scripted testing is better done in your main dev language
Source: I have been working with Postman a lot this week. It's bad.
Like Postman, without all the bullshit. Which is to say, like a few-versions-ago Postman.
The advantage of Bruno (and the VS code plugin) that storing data in "just a file" is friendly to keeping a local copy, and also to sharing, reviewing and versioning with existing tools (git).
How does insomnia handle that?
In general I like Postman because of the ergonomics, not because it does something that more primordial tools can't. Makes debugging easier.
Yes! It's supports functional BDD-style tests, "mock servers" (really "fake" servers), a walled garden for saving and sharing data (to frustrate engineering managers and ops folks), perf tests (soon? now?), contract testing, and more. Did I mention that it's a GUI tool, so it must be "easier?"
They're trying to become a one-stop-shop for dumbing down web dev. Ideally, inside the walled garden. That has a certain appeal to management.
We have lots of unit, integration, and live canary testing, the postman usage was only during oncall/operations, so we weren’t using any of those neat features.
2. It doesn't actually replace a USB drive. Most people I know e-mail files to themselves or host them somewhere online to be able to perform presentations, but they still carry a USB drive in case there are connectivity problems. This does not solve the connectivity issue.
3. It does not seem very "viral" or income-generating. I know this is premature at this point, but without charging users for the service, is it reasonable to expect to make money off of this?
"Dropbox’s stock falls as it struggles to squeeze more revenue from existing customers" - https://siliconangle.com/2023/02/16/dropboxs-stock-falls-str...
Setting valuation aside, clearly there is some non-negative value prop for some set of users who don't want to build everything themselves (as evidenced by the fact that they have a non-trivial number of paying users).
"I am become pedant, destroyer of discourse."
Here's the comment if you're curious.
> As a [type of tech person], [product] is completely unnecessary as you can easily build a similar system yourself by simply combining [tool A] with [tool B] and [tool C]. It's so simple!
I'm not disagreeing with your point, but it misses the point that often the benefit of using a particular product is that you don't have to build it yourself, not to mention the other benefits you could get from using the exact same system as somebody else (e.g. sharing Postman collections).
Also I can stitch together different APIs by taking response from one API and pass it as request to other APIs. For instance I invoke login API, grab session ID from the response and pass it to other APIs which are behind login.
Of course one could do all of this in curl but it’ll be cumbersome. And yes, if all you want to do is invoke a single API then Postman is definitely an overkill.
It’s like saying web browser doesn’t do anything that curl doesn’t. It’s technically true but not really.
But it’s also technically incorrect. Even on request level, not speaking about all the things on top that Postman does, Postman also does grpc and graphql (edit: and websocket), which curl doesn’t.
Once you start adding all the features, it’s very hard to replace Postman with anything.
Postman can even export a GraphQL request to a curl command.
Yes, a cursory examination of postman would tell you that.
Maybe when they actually deliver parcels like an actual postman? Fedex is worth more.
edit: check out her interviews! crazy impressive https://www.listennotes.com/top-podcasts/jean-yang/
- Netsil [0] (acquired by Nutanix) - Academic spin-out from UPenn
- Pixie Labs [1] (acquired by New Relic)
[0]: https://thenewstack.io/netsil-visualizes-performance-microse...
[1]: https://techcrunch.com/2020/12/10/new-relic-acquires-kuberne...
In addition, these techniques have been prone to event misses (partial reconstruction) in case the ring buffer overflows and are ineffective in the face of SSL traffic.
https://devboard.gitsense.com/akitasoftware
The org has had 304 contributors over the course of the org's history though.
Full Disclosure: This is my tool
After raising money, they bloated the app with confusing, half-baked features, and whoever was in charge of their usability really botched it. It's unusable now and I used to dread opening it up.
I switched to Firecamp[1] and have been very happy with it.
That said, it's not worth paying more for than GitHub per seat for my team, so we do without.
To stay unbiased, here's a list of other web based postman alternatives:
- https://hoppscotch.io (this is the most popular one)
In the next 1-2 months you will see the biggest amount of improvements the Insomnia application has ever seen, since here at Kong we are pouring tons of resources into it.
That said, powerful observability tools (Solarwinds anyone?) are really hard to make persistently secure, especially as they grow in their cumulative observations. I'm not really sure there exists a turn-key approach to secure these systems at scale without oppressive overhead or a near perfect team.
I wouldn't be surprised to hear about a mis-configured security group on an Akita instance caused a sev1 within 5 years.
Definitely not unique to either Akita or Postman but always worth discussing IMO
Congrats to Jean and the Akita team!
Postman on my Mac is now broken because a big JSON (almost 1mb) is stored on one of the opened tabs which makes it freezes on launch. I'll have to purge everything and install it again, thankfully that tab isn't saved in my profile so it won't be there...
I think fixing that is more important that acquiring new companies.
Maybe it's a Postman on Mac thing, I've never used it on any other platform.
Also... I think you're being a little pedantic here. Bash scripts are the automation already available to every terminal users. Their use is implied.