I'd say Valve is to software as Apple is to design, but there are probably far better examples. And I'm not a gamer (but I do use JIRA and SourceTree).
My only beef with Crucible is that it doesn't allow me to work in the way I want to work. I want to initiate reviews of another person's code. But with their whole moderator/author/reviewer workflow, I can't. I am the moderator and reviewer, the other person is the author, and the software won't allow it.
Other than that, I've been very happy with Crucible for code reviews.
We (25 people company) use Confluence, and we all truly hate it. The design is attrocious. It's ugly beyond the pinnacle of ugliness. We tried to use some custom styling to ease the pain, but Confluence uses so many badly designed elements that we had to capitulate.
If this incredibly crappy product sells itself, then it's only because there is nothing better. If you don't know what your next company should do - please consider creating a better corporate wiki. Confluence is truly an abomination.
There is probably already a company out there making awesome corporate wiki's and they are growing every year, but of course by the time you get it it will be out of date and there will some some new little company...
Amen. Last time I used it, it was a mess beyond belief. Code intermingling with presentation layer. Dependencies upon dependencies. A case study of bad software engineering.
Do you think there's demand for a "corporate wiki"? I've been toying with the idea of that on the side for months, but I'm unsure of the demand for it. It seems like wikis are something that people don't enjoy using and don't have a need for. What are your experiences? Where does Confluence fall short?
To sell a successful corporate wiki, you need hundreds of plug-ins to handle all those special cases that each company inevitably has. That would be your biggest stumbling block, because creating those takes time, and that's where Confluence trumps all others.
Where it falls short? Basically everywhere else. There is no taste.
If you want our business, create a wiki that puts the content front and center, and remove everything else by default. Then let us add our 3 extension wishes, while still staying minimalist. It's not that difficult.
Unfortunately, it seems that either the chief software architect at Confluence is not up to the task, or that the company values quick crappy changes in favor of good design.
I appreciate Atlassian tools. I'm a long time JIRA user, and I've been a Confluence and Bamboo user in the past. They're all decent tools. I have no problem with a nice article about a pretty neat company that makes pretty neat enterprise tools (an industry riddled with stuff that's mostly worse).
It's really just the article's title that's completely fucking bananas. Apple's relationship to design is profound. Write a book that covers the last half century of consumer product design, and Apple deserves a chapter. Do the same for Software and does Atlassian get a chapter? Probly not. Ironically, Apple probly does.
Write a title that includes a reference to Apple, then write some body text without doing to much research, keep the contributor count low so you don't have to spend time editing, then post on well known website. profit.
Seriously though there is nothing in that article that makes me think it is based on any type of truth or real world user feedback.
It varies wildly from product to product. When I worked there, every product was written in Java. From what I hear, these days, it's a little different.
This article is shockingly bad, from the very first sentence:
> There are only 3 enterprise-grade technology products I’ve ever seen that sell themselves. Two of them are from Apple
Huh? What "enterprise-grade" Apple product? XServe? Apple is nowhere in the enterprise, and thank god for that.
Here's some enterprise software that sells itself: Exchange. BES (well, used to). ProCurve. DRAC. Symantec Ghost. Did you see iTunes Group Policy Server anywhere in that list?
The article starts off stupid and only goes downhill from there. This guy doesn't know what the hell he's talking about.
The real story should be how they put the Confluence back-ups on the same raid disk as everything else. Yeah, thankfully we had back-ups when their server went down. Or maybe how sub-tasks on JIRA can't have sub-tasks. Or that any other thing I've seen that tries to compete with it looks better/feels better/gives a better user experience. Or... should I continue?
They recently removed wiki formatting from their wiki in JIRA. If you want to edit a page you have to use their in-browser rich-text editor.
This (on top of a bunch of other friction I have with JIRA) makes it clear to me that JIRA is not a product they're building for hackers. It must be for program managers, or for check writers at enterprise companies.
JIRA (And other Atlassian tools) are notoriously bad at scaling. In a previous life, I spent a week pulling a 2,000 user JIRA/Confluence/Bamboo stack off to my then employer's stack -- the site couldn't handle more than 40 concurrent users, and this was when the software was on a beefy box with tomcat and a high-perf MySQL Instance. Saying that Atlassian is enterprise is a stretch -- it might be great for small teams (personally, I'm a fan of request tracker and mediawiki), but, I'd never stick it in the enterprise.
25 comments
[ 5.1 ms ] story [ 109 ms ] threadI'd say Valve is to software as Apple is to design, but there are probably far better examples. And I'm not a gamer (but I do use JIRA and SourceTree).
Other than that, I've been very happy with Crucible for code reviews.
If this incredibly crappy product sells itself, then it's only because there is nothing better. If you don't know what your next company should do - please consider creating a better corporate wiki. Confluence is truly an abomination.
I don't hate Bitbucket, but that's an acquisition product, not a home grown product.
Where it falls short? Basically everywhere else. There is no taste.
If you want our business, create a wiki that puts the content front and center, and remove everything else by default. Then let us add our 3 extension wishes, while still staying minimalist. It's not that difficult.
Unfortunately, it seems that either the chief software architect at Confluence is not up to the task, or that the company values quick crappy changes in favor of good design.
Good luck with your endeavor!
It's really just the article's title that's completely fucking bananas. Apple's relationship to design is profound. Write a book that covers the last half century of consumer product design, and Apple deserves a chapter. Do the same for Software and does Atlassian get a chapter? Probly not. Ironically, Apple probly does.
You could say that a lot about Forbes online content these days.
Seriously though there is nothing in that article that makes me think it is based on any type of truth or real world user feedback.
> There are only 3 enterprise-grade technology products I’ve ever seen that sell themselves. Two of them are from Apple
Huh? What "enterprise-grade" Apple product? XServe? Apple is nowhere in the enterprise, and thank god for that.
Here's some enterprise software that sells itself: Exchange. BES (well, used to). ProCurve. DRAC. Symantec Ghost. Did you see iTunes Group Policy Server anywhere in that list?
The article starts off stupid and only goes downhill from there. This guy doesn't know what the hell he's talking about.
This (on top of a bunch of other friction I have with JIRA) makes it clear to me that JIRA is not a product they're building for hackers. It must be for program managers, or for check writers at enterprise companies.