Show HN: See what the best startups in the world are using
leanstack.io
leanstack.io
Companies like Twitter, eBay, GitHub, Mozilla, Netflix and Atlassian are past the startup stage by now. Many of them have been market leaders for years.
Calling Intuit a startup isn't even stretching it anymore. They're 30 years old and have a revenue of over $4B.
I realize HN thinks startups are sexy, but these have no business being labeled startups. If Intuit is a startup, what does it take to be a called a mature company?
Replacing "startups" in the title with "(technology) companies" would be more accurate, but probably reduce traffic.
I'm sorry but on what planet is eBay a startup?
This comment is in the same category as the one PG talks about here: https://news.ycombinator.com/item?id=4693920
I don't want to pick on you because there might be a place for your comment but it's not at the top; we need to stop voting these to that position (I'm not encouraging down-votes).
Secondly, if it shouldn't be at the top, what should be? I think you are looking at spiff's comment in a vacuum and saying, "This is not platinum-quality comment material!". But there isn't any platinum-quality comment in this page of comments yet. You can't upvote the rightful comment to the top if it doesn't exist yet.
While it is not outright dismissal, it is dismissive in nature, and again, my primary issue is with voting, not necessarily the comment. I think we could still ask, quoting PG, "Yeah, we know that [Intuit isn't a startup]. But is that the most interesting thing one can say about this article?"
[and now it's getting down-votes after your comment despite PG making it clear time and time again that this is a major problem for HN]
-----------
For the time being, yes, it is. And given that it wasn't an article being linked to, but a site that included ebay and intuit under the banner of "startups", yes, this might be the most interesting point of the site people want to discuss.
It's a hypothetical not a question of the present state of things but either way, I disagree. Rachel's comment [1] (I understand it wasn't present when I made my first comment), while terse, is more interesting, more likely to foster interesting discussion and doesn't have the dreaded HN "middlebrow dismissal". And I have no doubt HN can come up with much, much better. The point being, it became a "magnet for upvotes" and nothing better ever had a chance to reach the top to encourage more interesting discussion.
> but a site that included ebay and intuit under the banner of "startups", yes, this might be the most interesting point of the site people want to discuss.
That's unfortunate; I've found HN to be a place to find great tangentially related discussion to otherwise uninteresting posts. That doesn't happen by accident; we need to foster it.
it became a "magnet for upvotes" and nothing better ever had a chance to reach the top to encourage more interesting discussion.
This is an interesting way to think of it. I think we could agree that the most valuable comments will not necessarily appear until the article has aged a bit. Clearly HN tries to combat this by moving certain new comments to the top for a little while, but perhaps something more dramatic is needed. Perhaps for a certain period in the thread's life, for example, distribute comments with positive scores using some random seed such that higher rated comments will generally be near the top to avoid wading through a morass of poor quality, but any one (positive-scored) comment might be at the top for any given refresh.
P.S. I don't think you deserve the downvotes you are getting. I obviously don't exactly agree with you, which is why we are having this discussion, but I think you raise valid concerns and make them in a reasoned and articulate manner.
It's very easy to make a quick, correct, point that avoids discussing any of the content of the article. That comment will get upvotes, and will be top of the list, and will continue to get upvotes, and thus will stay near the top of the list. It self-perpetuates.
For me, I think the issues that make discussions go "downhill" are less to do with criticism and much more to do with prevalence of group-think. I don't think the prevailing wisdom that pointing out fundamental errors of facts of faulty arguments is a bad thing, even if they are not the main point of the article. I think there is edifying value in such criticism. Now, it is true that an article may have value beyond it's minor or major inherent fallacies and errors. Criticism is not censorship and it does not prevent others from reading and discussing other parts of an article if they are interesting and fodder for good conversation. Nevertheless, I don't think being overly generous or charitable necessarily the best method or should be enforced via an algorithm or community brow beating. No one wants inane Reddit conversations but I don't think most want sterile group-think or hug sessions either.
HN has all kinds of users with different backgrounds and some users are very knowledgeable and critical and we can learn a lot of have great discussion regarding the criticism as the more traditionally positive style conversations. If we eliminated this sort of discussion from HN, I think it would make things much worse than better, Valley group-think would rule the day. One thing I think users that come to HN expect is good information either in the articles or to find the proper information in the comments. If we let erroneous information stand because of we want to be nice or follow Atwood's "ruthlessly civil" or whatever the hell his ideology is, I think we would less a great benefit of HN discussions and users, particularly less experienced ones, would be worse off. If I'm new, and I see an HN submission that says eBay and Netflix are startups and that clearly is a significant part of the article and as well there are many comments on HN, but with no contention of that issue anywhere to be found, I may well think that the community consensus is that eBay is a startup.
On other articles that get submitted, I may conclude that the smart people on HN think all manner of pseudoscience and homeopathy are generally accepted, if I wasn't aware of there being new a impetus to avoid criticism.
I personally don't want what is suggested by this random author: http://www.modernperlbooks.com/mt/2013/03/mrs-feynmans-advic...
>Is your problem that the hearty self congratulation mutual admiration society on Hacker News (now there's a middlebrow dismissal!) thinks that something you like isn't as new and shiny as something you don't care about?
(oddly enough I think that article is relevant to the original topic) I think that comment is somewhat unfair to HN but if we are not careful and are to act to hastily toward criticism we may just end up with a "mutual admiration society." Let's no do that. :)
-------
2 decades after the IPO should be the cutoff point. After that, you're "mature"; before that, you're still considered a 'scrappy startup' which might fold at any moment without the day to day pivoting of the charismatic founders in their garage. ;)
for example, I know Instagram uses python, redis and Postgres. Pinterest python/django but moving to flask and using MySQL. But I'm curious to know about other sites.
Also, a graph with dependencies between these companies would be cool to see. Security companies might want to give the node with most edges a call or two.
It's tabular data. Put it in a plain old HTML table so we can make sense of it.
Is this something you'd find useful?
The ability to view a company's stack changes over time might prove more useful. For example, what stack did Company X use to get to 10k users, 100k users etc.
Given enough accurate data then it may be possible for a visitor to predict any performance bottlenecks that a technology may enforce over time and what technologies to use to mitigate said problems.
Truly good developers will know to discard poor tools in favor of better ones, and will be eager to do so. They don't have emotional attachments to, say, Ruby on Rails or MongoDB, or some other piece of software or service. Once the tool becomes insufficient, it's gone. In many cases, this is obvious right away, so the tool isn't even used in the first place.
In all seriousness, I would PAY to see this. I would pay extra if you add hosted graph widgets that empirically conclude the largest reciprocal relation between valuation and an IIS/MSSQL stack.
I will take a gander however and assume someone didn't like what was presumably a rhetorical and pretentious response.
I do though agree. As to your first response, a quick right click will show that the site is made with Wordpress. The lack of formatting for the stack grids suggests it MAY be a plugin from which this is being retrieved; nevertheless with the choice of Wordpress, I am doubtful.
Perhaps I don't understand the hn culture, is suggesting ways to do something more efficiently against the rules of Show HN? And if whoever keeps downvoting me wants to discourage me from making similar comments, how about explaining what is wrong with my comment? Instead of anonymously downvoting it.
There's got to be a better way to categorize the technologies used. Perhaps separating the core web request stack and from the support components would be a start. Some aggregate information across all of the companies would be useful as well.
It would just be a fun fishing expeditions with the data.
Fir example, Ebay ?
While in New York I went to some networking events where people said 'I work for a startup', 'How long have you guys been around?', '4 years', 'That doesn't really sound like a startup.', 'I know, but we FEEL like a startup.'... A*holes.
It is like saying that your car or house is new after 10 years. Or saying your shoes are new when they are 6 months old.
The word is pretty self-explanatory, it is not that hard people.
Sure, many others may be startups, perhaps even most. But since when do we generalize away ~30% of some population. That's not very scientific. I can't think of many scenario's where this would be acceptable.
BTW -- Unicode fail at: "99designs is the world’s largest..." and other places. Check your page's UTF-8 support.
Everyone, always: If you make websites on a super fast connection, get a proxy (or VM) to simulate a slow connection...! You will learn a lot of things which will become easy habits quickly, and then you can turn off the proxy (turn it on again occasionally).