17 comments
Comment actions Permalink

Comments-

1)Thumbs up on release promotion, build comments, build labels, blinking tests, custom content, server-side delayed commit, customize rollup status.
2)Maximal thumbs up on notification improvements, rss especially! (but sadly no Mac dashboard widget)
3)Statistics has enormous coolness potential
4)Looks like code review functionality got pushed off to 4.0. Probably sensible.

All in all, very nice. Focus looks to be on completeness and "enterprise class" functionality, rather than pushing boundaries. Lot of polish, here.

--Dave Griffith

0
Comment actions Permalink

Hi Kirill,

I'm sure you've been too busy on TeamCity to read the ReSharper threads related to pricing/upgrades regarding the soon-to-be released 3.0 version. I'm a bit miffed that after only a few months of paying for a version of JetBrains software, that I'll have to pay for an upgrade.

I'm really very interested in TeamCity, but there's no way I'm going to get caught paying for it twice, as JetBrains would like me to do for ReSharper. As I said, it looks like an interesting product, but I won't even download the eval because I'm not willing to even consider the notion of getting screwed again (as I see it anyway.)

Cheers,
Giorgio Galante

0
Comment actions Permalink

Usually when JetBrains publishes a roadmap, it means the actual product is something like nine months to a year away. This roadmap feels about that long, based on past velocity of TeamCity development. I can personally gaurantee you that if you buy a license for TeamCity, you'll have recouped your investment in far less time than that. (Same with IDEA or Resharper, for that matter).

--Dave Griffith

0
Comment actions Permalink

"I can personally gaurantee you that if you buy a license for TeamCity, you'll have recouped your investment in far less time than that."

That answer doesn't fly when it comes to asking for a budget for a tool (and then coming back a few months later asking for more money for an upgrade.)

Check the ReSharper threads and you'll see that some folks have actually stopped using Resharper in their corporate environment because they weren't willing to pay for an upgrade so soon after purchasing multiple copies.

0
Comment actions Permalink

Check the ReSharper threads and you'll see that some folks have actually stopped using
Resharper in their corporate environment because they weren't willing to
pay for an upgrade so soon after purchasing multiple copies.

Maybe that was the case with ReSharper but it's for sure not the case with Jetbrains Java based
products.

We follow these lists since more than 6 years and if you do a search on the announcements lists
you will see that it's exactly like Dave said.

Ahmed.

0
Comment actions Permalink

I totally agree on this point. This is true for all Jetbrains products that are all wonderful.
We always miss the next features that make the products greater (not always the case with other companies).

Even in small companies it's not easy to manage this, management would prefer that we use free product like Eclipse or CruiseControl, it's cheaper. We already have to prove that the ROI is real, easy for (hard)core developpers but harder for mainstream developpers and managers.

I and my team, are not ready to choose another products because Jetbrains rocks. My company is not ready to pay a one shot 10000$ to get all future releases ( I don't have the budget :( ).
I understand that Jetbrains needs money to grow, and I want it grows and provides more powerful products. So maybe the marketing should work on a different pricing policy (some kind of loan)

Just my 2 cents

0
Comment actions Permalink

>That answer doesn't fly when it comes to asking for a budget for a tool (and then coming back a few months later asking for more money for an upgrade.)

My conservative observations have TeamCity saving our developers something like 20 minutes per day in workflow simplifications, specifically around remote run and deferred commit. There are other benefits, but let's not complicate things. That 20 minutes doesn't include the baseline benefits of continuous integration which could be gained by going with an open-source solution. At 20 minutes per day, and an extremely conservative estimate of $100/hour for all-in developer cost, that has TeamCity paying for itself after nine days of use. Everything after that is profit. If your management isn't smart enough to make that trade, they aren't smart enough to be managing software developers in the first place, and you really should consider polishing your resume.

Incidently, IDEA saves much more, because it can actual decrease bug finding/fixing time, and decrease bug injection rate via automated refactoring. I'd put IDEA productivity bumps on the order of two hours per day over pre-IDEA IDEs, and still at least a forty-five minutes per-day over the current "me too!" freeware options. IDEA licenses typically earn out in less than a week.

I'm sorry to hear you've been having stability issues with ReSharper. I haven't experienced such, but it's certainly not unheard of with technologically aggressive products. Thinking that you are being ripped off because of it is just foolish.

--Dave Griffith

0
Comment actions Permalink

Dave,

I'm not questioning the value of TeamCity (or Resharper for that matter.) I'm questioning the inconsistent upgrade policies by JetBrains. They're not sure if they're going to charge existing customers for the 3.0 release. That makes me want to delay evaluating the product until the upgrade pricing policy is established.

I also don't think I'm necessarily being ripped off. All I ask for is some consistency (buyers of Resharper 1.0 have received free upgrades up to version 2.5.2 - great for them. I bought Resharper @ v 2.5 and I recently learned I'll have to pay for the upgrade to 3.0.) I don't think it's unreasonable to ask for consistency to that end and I won't take a chance on this happening again with TeamCity 2.1.

In other words, JetBrains will have to wait to receive our order. If they don't mind, so be it. This pricing strategy will only lead to delayed purchases or the use of other comparable products.

0
Comment actions Permalink

That strategy may be odd, but is not actually inconsistent. For both IDEA and TeamCity, 1.0 customers got the 2.0 cycle for free. After 2.0, each IDEA release was a separate license. I'd be very surprised if TeamCity 3.0 doesn't require a separate license. If JetBrains says they're not sure, it most likely just means that Alex or Sergei hasn't gotten around to signing off on the terms. Since the Benares release isn't likely to be for quite a while, that doesn't seem unreasonable.

It should be said that such a policy is not uncommon for the industry as a whole. It gives "pioneers" an incentive, without overly complicating later licensing or giving away too much for free. I've never heard of anyone who has a default "if you buy version X, you get version X+1 for free".

--Dave Griffith

0
Comment actions Permalink

Again, the blog states "The most optimistic date for the release is November, 1", which says to me, anyone that purchases TeamCity 3.0, 3 months prior to the release date will likely be eligible for a "free" upgrade and folks who buy it say today, will not likely get it for free - maybe.

That last "maybe"/unknown is the part that bugs me. Most of the big software players announce their "grandfather-ing" upgrade policies/dates beforehand. i.e. Adobe:

"From March 27, 2007* through August 28, 2007 you can buy current releases
of Adobe Production Studio Premium, Adobe After Effects 7.0 Professional, or
Adobe Premiere Pro 2.0 software and get a complimentary upgrade to the
new CS3 version of these products or their equivalent when they ship. You will
also be able to upgrade from Production Studio Premium to Adobe Master
Collection CS3 at the regular upgrade price."

Printed in the USA. 02/26/07"

0
Comment actions Permalink

Just so I understand. You're going to delay a purchase for something like six months, in order to avoid paying an extra license fee, on a product that would pay for itself via increased developer productivity in a time frame of less than one month?

It's been a while since I took economics, but when I put it that way, it sounds kind of, you know, stupid. What am I missing?

--Dave Griffith

0
Comment actions Permalink

Not everyone has CNN's budget.

Your previously stated cost $100/hr per developer is extremely high in my town. So the time to recoup the cost of having to buy TeamCity now and again in 6 months is simply not 1 month or less. Sure that's not your problem or JetBrains' problem.

If JetBrains would just put a stake in the ground now and say, 2.1 buyers will (or will not) receive 3.0 for free - at least I can make a better/more informed decision. It seems like they're playing games. Customers don't like games. I'm glad you work for a huge company that can procure these wares easily - I do not and cannot - which does not make me stupid. You'd have done well to preview your last post - you sounded like an arrogant prick.

0
Comment actions Permalink

Not everyone has CNN's budget.

All the more reason not to waste money on misguided purchasing policies, right?

Sorry if I offended, but I still really don't understand how your plan is anything other than profit-minimizing. Even at Indian rates, scrimping on tools as cheap and productivity-enhancing as JetBrains provides just doesn't make any damn sense. The numbers just don't work. The end result is that you end up having to hire more developers for the same work, or dropping work on the floor, because you couldn't pony up a three-figure license fee. That's madness. Unfortunately, it's a sort of madness that is endemic in our industry. Finding yourself at a company where such bad decisions have been embedded as policy, you're really better off looking for better work.

FWIW, I own my own IDEA license, purchased out-of-pocket. My employer would do so, but I do open-source work on the side, and prefer the control that owning my dev environment provides. The license fee is trivial compared to the value provided. While your employer may not have CNN's budget, their budget is surely higher than mine.

--Dave Griffith

0
Comment actions Permalink

>Sorry if I offended, but I still really don't understand how your plan is anything other than profit-minimizing

No worries - but it's not my plan ;) In the market I'm in, that's the reality at most every shop. Does it make sense? Nope - I certainly agree with you there.

I own my own Resharper license and am considering buying my own TeamCity license regardless of my employers policies. Given that fact, I personally don't want to spend more of my money than I have to (if I were given the option of spending someone else's money, I wouldn't care so much.) I'm sure you can understand that - and that's what it ultimately comes down to for me.

0
Comment actions Permalink

Hi KIR,

Roadmap looks good (especially glad to see the Notification enhancements
- perhaps I will finally be able to get rid of a whole bunch of my mail
filters)

I would also love to see better support for Functional Tests;
specifically, we have a few JUnit tests that test the state of our
system against a baseline. It is quite a long procedure to generate the
baseline, hence we only do it once a day; so for us it would be really
helpful if we could assign developer responsibility on a per-test basis
rather than on a per-build basis (
http://www.jetbrains.net/jira/browse/TW-1319 ); and also if we could
have a third test state which is neither pass nor fail, but "fail but
explained" (a bit like the current "Fixed" I suppose)
http://www.jetbrains.net/jira/browse/TW-2829

Also it would be great if an lpr build performed the same steps as it
does in Idea (i.e. compiling JSP, running RMIC if necessary, etc.) but I
guess this is maybe for the BuildDesk people ...

Thanks,
R

0
Comment actions Permalink

I'd like to see:

1) Better support for writing plugins. API documentation, binary JARs, etc.

2) More plugin points through which to extend the web app, add links to pages or into tables. E.g. I'd like to add a link from a build type row on the front page to a build monitor for that build type.

3) A test grid. Farm tests out to agents, not entire builds, so that long running system tests can be run in parallel in isolated environments.

0
Comment actions Permalink

Thumbs up for the workflow support! Along with this would go user roles, e.g., developers push the "Build a release" button, QA pushes the "Deploy to production" button.

Right now, we are happily using TC for the "Build a release", right next to the "continuous build", but it stretches the current TC model a bit to set up a "Deploy to production" button.

If you get much into the development workflow space, TC will need to be comfortable swapping ticket info with issue trackers, e.g., JIRA. We're thinking about driving TC builds from a JIRA workflow, but JIRA isn't so well set up for this out of the box, and TC feels like a better one-stop shop for build management.

Cheers,
-chris

Advanced build promotion action with possibility to publish it to external location, pin, tag, prepare changelog, and execute arbitrary script

0

Please sign in to leave a comment.