Disqus - Latest Comments for nkohari

Re: I'm not nearly the man I used to be

Nate Kohari — Thu, 17 Jul 2014 12:41:09 -0000

Congratulations David!

Re: Snapchat's 3 Billion Dollar Blunder

Nate Kohari — Fri, 15 Nov 2013 10:19:21 -0000

There are a tremendous amount of startups that are profitable. Why is it insane to call those companies startups, and what's your point exactly?

Re: Leadership and Communication

Nate Kohari — Fri, 01 Nov 2013 06:55:14 -0000

Sure -- I think the same principles apply, but in a flat organization it may end up being more of a shared responsibility to allocate and match work. That said, even if their title doesn't reflect it, more than likely someone in the group will end up acting as a leader.

Re: On Nate’s Departure

Nate Kohari — Wed, 16 May 2012 15:17:42 -0000

Thanks for the kind words. I hope you guys already know this, but for the record you're absolutely, hands-down, the most amazing group of people I've ever had the pleasure of working with. I'm sorry that I wasn't allowed the opportunity to see through some of the great things that we've been building, but I'll always think back fondly on my time working with all of you. Best of luck in the future, and I'm sure you'll all do excellent things, whether with Rally or not.

Re: Happy New Year!

Nate Kohari — Mon, 14 Feb 2011 09:49:44 -0000

We're doing some meta-work right now out of necessity more than anything else, but trust me when I say all of the goals the work supports are customer-focused.

Re: Happy New Year!

Nate Kohari — Thu, 20 Jan 2011 15:45:39 -0000

Actually, that's part of the reason why we like CoffeeScript so much -- it's what JavaScript's syntax should have been all along. Don't get me wrong, I love JavaScript, but it's missing a few things. Coffee provides an inheritance model and reduces the amount of ceremony you need to do simple tasks.

Re: Updates to Story Cards

Nate Kohari — Tue, 14 Sep 2010 11:29:57 -0000

Sorry about that -- you must have caught us at exactly the right (wrong?) time.

Re: Filtering Performance Metrics

Nate Kohari — Fri, 20 Aug 2010 10:51:22 -0000

I can see what you mean. The charting library draws a shadow effect under the line, which can get in the way on some charts. We've tweaked the display of the cumulative flow graph to remove this shadow and reduced the weight of the line, so the charts should be a bit clearer.

Re: Introducing Filtering

Nate Kohari — Thu, 22 Jul 2010 21:03:39 -0000

I understand your perspective, and I appreciate that our software is able to evoke such a strong response in our users!

I don't believe I ignored your use case in my reply -- it's a reasonable way of using the software. That being said, there are two distinct problems that I think you're conflating: *filtering*, which produces a limited view of the board based on well-defined criteria, and *search*, in which you're looking for something that you can't find, based on what you know about it. The ideas are similar but not identical, and while we do eventually intend to solve them both, only the former is solved in this release.

We certainly understand the Agile concept of releasing early and often, and I believe the product's history has shown that we practice it pretty well. We also believe strongly in the Lean idea of the minimally-marketable feature. In filtering's case, the shortest path to release that provided the highest value to our users was to implement the expression language as we did.

We could have solved both problems at once, but from a development standpoint, it would have been significantly more expensive. In order to implement full-text search without degrading performance, we would have had to integrate a search engine into AgileZen. Since that would have required quite a bit of additional engineering effort, we decided the expression language allowed a great deal of power, while not significantly degrading the user experience.

In the end, by implementing the feature in this way, we got the most often-requested feature to market very quickly, while still leaving the door open to expand it in the future by adding full-text search capabilities. We've provided a solution only to the filtering problem (which is a minimally-marketable feature), while leaving the search problem to be solved by a future release.

Thanks again for sharing your opinion! So far we've had a very positive reaction to the release, but I'm always interested in hearing criticisms as well as praise. :) I'd encourage you to continue to use filtering for awhile. After you get accustomed to the filter expressions, I think you'll agree that it's really not that difficult to use.

Re: Introducing Filtering

Nate Kohari — Thu, 22 Jul 2010 13:30:02 -0000

Thanks for the feedback! You have a very valid point. When we first started talking about implementing filters, we started out with the idea of using a full-text search, and it grew into the key/value expression language. There are two reasons why we went with the expression language:

1. Performance. It's very expensive to search for a keyword across all the different fields of a story. That can be mitigated by introducing a search engine, but we decided that wasn't worth the additional effort yet. This will likely change in the future.

2. False positives. When you're looking for a webpage via Google, it makes sense to use a full-text search. However, when you're looking through items that all have the same structure (like stories in AgileZen), it's pretty rare that you'd actually want to look for *every* story involving email. Instead, it's more likely you'd have tagged everything with an "email" tag. The expression language requires that you understand what you're looking for a bit more, but in return it makes your search quite a bit more powerful.

In the end, I don't think the expression language is really too complex for anyone to use it for the simple cases, and as you become more accustomed to it you can use it for the more advanced ones. Plus, the ability to save filters allows power users to create complex filters and share them with the rest of the team, who may or may not be power users.

Re: self serve advertising on reddit is now open for business!

Nate Kohari — Wed, 09 Dec 2009 17:19:26 -0000

We'd like to give this a shot for our startup, but I was confused about the timing of the campaign. I'd like to run an ad only Monday-Friday, but if I choose to run an ad from Monday 12/14 to Friday 12/18, it says my bid is only for four days. I take it the dates aren't inclusive, but which way does it go?

Re: The Next Step

Nate Kohari — Fri, 20 Nov 2009 16:36:31 -0000

Bill: Somehow I missed that when I was writing the blog post. I updated the post to add information about the new WIP limit system. Thanks for pointing that out!

Re: Firefox Add-On: MeasureIt

Nate Kohari — Mon, 09 Nov 2009 09:26:37 -0000

MeasureIt looks pretty cool. For individual elements, you can use the Firebug Layout display to see the dimensions (click HTML and then it's on the left-hand side). And for images, you can right-click and choose Properties to see the size. Just another way of doing the same thing! :)

Re: 18: Talking with Jeremy Miller about Alt.Net

Nate Kohari — Wed, 29 Jul 2009 13:36:22 -0000

@Bailey: I think you mean "kanban" ... it's a lightweight process based on ideas from the Toyota Production System. There are some good resources on the web about its application to software. And, if you want an example of a kanban board, you can check out Zen, which is our web-based project management system geared around kanban. </shameless_plug> :)

Re: Eating our own dog food

Nate Kohari — Fri, 24 Jul 2009 14:22:41 -0000

The size field is editable via the story details screen, in the upper left under the story reference number.

Re: The Perfect Margarita : Infozerk Inc.

Nate Kohari — Mon, 11 May 2009 10:40:41 -0000

I had to try this. Works great with Cointreau instead of Grand Marnier also. It's almost like a martini but with tequila.

Re: "Don't be a plumber" is horrible advice.

Nate Kohari — Mon, 20 Apr 2009 10:04:14 -0000

Leon, thanks for the kind words, and for putting up with my best Mario impressions. Like everything else, the debate can't be waged in black and white -- I believe it was the Byrds who said, there's a time for plumbing and a time for user-focused work. :D

Re: Patterns, Purists, and Sinkholes : Rob Conery

Nate Kohari — Thu, 12 Feb 2009 09:55:10 -0000

This is an absolutely fantastic analogy. Thank you for this comment, you made my day. :)

Re: McHypocrisy

Nate Kohari — Wed, 22 Oct 2008 10:10:58 -0000

Hey, what's wrong with Akron, Ohio? :) Seriously, though, your blog has become one of my favorite reads lately. Keep it coming.

Re: Comparing .NET DI (IoC) Frameworks, Part 2

Nate Kohari — Mon, 08 Sep 2008 22:22:39 -0000

Fantastic review! Thanks for writing it. By the way, the problem with the ExtensionAttribute has been fixed in the latest trunk build of Ninject, which will be released as 1.5 soon.

Re: Context Variables in Ninject

Nate Kohari — Thu, 13 Mar 2008 22:09:48 -0000

Actually, I'll add some case-insensitive comparison stuff to the StringConditionBuilder. It probably should be there anyway. :) Glad this will help you out, let me know if you come up with any other creative uses for Ninject!

Re: Ninject Lives!

Nate Kohari — Wed, 12 Mar 2008 20:52:31 -0000

I was considering using Beanstalk for keeping some of my other projects (which are too messy to release as legitimate open-source). Unfortunately it looks like the pricing model's not the greatest for that, anyway. I was originally attracted by the nice integrations they support -- for example, Twitter. :)

Re: Glitches

Nate Kohari — Tue, 15 Jan 2008 09:01:56 -0000

Malinski? Here? I'll blame ET for that. :) How's it going?

Re: Trying out Disqus

Nate Kohari — Wed, 02 Jan 2008 07:10:12 -0000

Posting a test comment. :)