In the blog post, you mention that these teams are adopting the Lean Startup philosophy of ship early and ship often. Does that mean that you've softened on the "never ever write any code without a spec" dogma from "the Joel Test"?
I've personally found that when you're doing tight iterations and continuous deployment, writing old-school spec documents feels, well, old-school.
If I can go straight from human conversation w/ whiteboard sketches to working, tested code running on production servers without creating intermediate written documents, I think I'm winning.
It's more complicated than that, and it's hard to answer here.
Basically -- if you're building something big and complicated and you should know exactly how it's going to work, a big written spec really helps.
If you're building something brand new, a startup, or a crazy idea that you need to test quickly and iterate, you should use lots of little specs. It's ok if the specs aren't so fancy... a detailed map of what goes on the screen and what everything does is probably fine. It's ideal if you write lots of small specs for each small feature ("user story") right before you implement them and roll them out.
I just don't like the idea of typing code when you haven't thought through what it's supposed to do (in a disciplined way). Doing specs in small quick chunks, then writing code, is probably the best approach for lean/agile startup teams that don't yet have customer/product fit.
Thanks for the response. I see where you're going from. I only brought it up because my development organization would fail that part of the Joel Test, where many totally dysfunctional organizations I have worked with would pass. Doesn't pass my initial smell test.
It's easy to see someone jumping in and writing and re-writing code when they haven't thought it through as wasteful. People have a harder time seeing that writing and re-writing specs can be similarly wasteful.
And reworking a whiteboard sketch for a feature takes a lot less time than rewriting the spec.
I think Joel's comment of "thinking in a disciplined way before typing code" hits the nail on the head. Different people can think in disciplined ways before typing code. Some people do that by writing specs, other people do that by writing tests, others do it at the whiteboard. There's no one right way to do it.
As someone who can't draw a thing, there's a phenomenon I've noticed that might apply. Every time I need to draw something out, I try to hold it clearly in my head, where it looks so complete, but of course it never works. My suspicion is that instead of actually seeing the details of what I want to draw in my head, my brain has filled them in with "here be details," which makes for a nice mental image but that doesn't really play so well on paper.
I also think the same thing happens when we write software with loose mental maps. Our brains fill in the gaps with "yadda yadda yadda," to make it feel complete, but really there's nothing there at all.
Practical suggestion that has worked for me is to use graph paper. Having the grid to align and separate things makes a huge difference. My non-graph-paper sketches look like something from a lunatic's diary in comparison.
I would be eager to try a whiteboard with a consistent faint grid to it.
+1. I can't draw a smiley face. Plain old graph paper has had a huge impact on my ability to put together satisfactory mockups. I have a pile of it sitting right next to me in fact. :-)
I would really like a tool that would allow a spec and development notes to evolve just as a program evolves. It should start out with being as simple and low-overhead to use as a few post-it notes, but as the projects gets bigger it should be possible to impose structure, like categorize notes into functional spec or requirement, user stories etc, prioritize etc.
I believe this is called literate programming (LP), which focuses on real documentation (in the broader sense, including specs, etc) rather than mere API docs. There are quite a lot of LP tools, many of them being independent from the language.
However, LP is not about tools. I've written LP programs that don't need any extra tool[1]. Rather, LP is about putting the documentation (spec, requirements, etc) at the highest level, and make the code a sublevel of that. This is very different from the traditional approaches where code is the highest level, or where documentation and code are separate.
[1] This worked because the build script was part of the literate program, too. Many "problems" with LP solve themselves if you just apply LP consequently.
Why not? For instance, feature proposals are just the previous phase of high-level requirements. And high-level requirements are just the previous phase of (or introduction into) the spec.
I'd think of it as a big documentation where some parts are complete and mature (down to the level of code), while other parts are still in draft (feature proposals).
Or change management: Of course everything should be under version control. And I guess that bug descriptions should still be part of a ticket system or at least the commit messages. But I never understood why feature proposals are discussed more in a discussion-based manner (ticket system) rather than a document-based manner (wiki, or documentation).
Unfortunately, I was never part of a development team which was seriously into LP, so my suggestions are only based on my understanding of LP, and small proof-of-concepts on my own.
I'd love to read about bigger real-world projects which apply LP, or which at least apply some other documentation-driven approach.
Nevertheless, I have done some experiments. Some time ago I submitted a small demonstration about "Self-Contained Literate Programming" to HN. It seems to have vanished, so I just resubmitted it. It would be great to get some feedback/discussion about that:
This actually sounds more like lean startup than unlike it. Yes lean startup encourages getting to testing your ideas quickly, but it also focuses on reducing process waste. If you have a mature product and enough data points suggesting major new feature, I think it'd be a waste to try and recollect those data points or build half of the features just for the sake of being lean even if in the past you have been mostly spot on with taking customer feedback and releasing successful major features based on them.
Hey Joel, this is terrific, and what's on my mind for quite a long time. None of the present kanban implementations are good, and I've been thinking of using node+socket+backbone to build one in my spare time. I've also downloaded the iPhone version but it lacks a lot of polish. TL;DR I like trello, and could I join your fabulous team?
Sniffing around the source, looks like a Backbone.js app -- cheers. I'd love to add it to the homepage as an example, if you want to email me a brief paragraph of description.
Edit For those poking around, check out the top-level "Models" namespace.
We do. Its hard to list every framework/library in there, so they're not all mentioned. Look for a future technical blog post that goes into more detail on what we used.
I user Chrome 14 and updates take over 5 secs to be synced across different tabs. Some kind of a bug or the backend app polls the db regularly to push new messages ?
We're scaling back a little to handle the usage spike from the initial launch, so some users will get polling even if their browsers support websockets. As the load stabilizes, we'll be switching them back on and you should see performance improve quite a bit.
Express, Async, Mongoose, Redis, Socket.io, CoffeeScript, Less, and Underscore, off the top of my head. We also use Node Inspector during development; that's a great little tool.
Really awesome. I'm using the same libraries right now in my own app and struggling hugely with Backbone. Would love to see some of the posts someone referenced a few posts up to explain how you fit it all together.
I am a bit curious about it's performance characteristics (e.g. number of node.js instances launched, method of load balancing, etc). Do you experience any problems with GC?
Trello is a collaboration tool that organizes your projects into boards. The site was built ground up using backbone.js for all models, views, and routes.
So I honestly don't get it. Is this a poor mans bug tracker? A possible re-invention of a bug tracker? (something wacky and different version of a bug tracker to see if it sticks?)
I have seen people mention project management a bunch, but the view really isn't about viewing what people are doing. In fact items that don't have people assigned to them still show up. The 10 foot view isn't even that good as they all squares and text. At a glance you can not tell what changed recently, what is late etc.
I had to really grimace when it showed the internal team that was using it and one of the stacks was "bugs" and it had the most number of items and was scrollable. ugg Does that scale to thousands of open bugs (or how about just 50)?
So either this is for all of those people who have never discovered the overview page of their bug tracker or maybe it is trying an experiment to see if the process of creating a bug tracker for a project is too difficult and here you just click "new project" and blam done and later on you export it to a real bug tracker... Maybe this is all just tricking users into using a bug tracker without them knowing?
In my view this looks like a great tool for agile workflow. A replacement for the board with sticky notes.
regarding the scale ... most if not all bug-trackers become a junk yard of bug reports. thousands of them. most never get fixed. lots never get looked at. This tool lets you store things that you are going to do in the near future. And the rest... lets deal with it when we get there.
I think if you can't remember it a month from now, it probably not important.
In any real project you can never deal with all the bugs and reports. so you usually deal with the most serious ones and the ones that get reported most often. the important ones go on todo right away, and the often reported ones do not need another place for them. people will keep reporting until you deal ;)
so this tool is effectively lets you plan your activities in the short and medium term. once you get to the 'next year' you will have other priorities and new things to do. there is no need to plan it now.
Im definitely going to try it. Currently we mostly work with Pivotal Tracker and also trying out Trajectory from Thoughtbot. It looks like Trello might be a decent contender in the space.
In looking at it, it looks much more generic than Pivotal Tracker (which I'm a current heavy user of and absolutely adore).
Pivotal Tracker is very focused on being an agile workflow tool. You enter a card, estimate it, start it, deliver it, and then it gets accepted or rejected. This is great for agile projects. Where it falls down is if you want to use it for non-development efforts, like say a marketing project, or building a new data center. Projects where that workflow doesn't really apply.
In my opinion, this is where Trello seems like it could shine. It's one tool that everyone can use to manage their work.
I don't see myself leaving Pivotal Tracker any time soon, but I will definitely keep an eye on Trello.
I am also a bit surprised we are seeing a kanban tool from fogcreek.
At work we have used fogbugz + the kanban plugin. Although the kanban plugin does not allow for the same level of fanciness when editing notes, the fact that it is integrated with fogbugz makes it more attractive.
I'll take a look at Trello, I suspect the UI may be nicer, especially when accessed from mobile devices.
to me it looks like just a simpler version of Fogbugz. I am guessing on the back end they re-used a lot of code. Fogbugz offers more features, but this is more accessible, easier to use, and more basic so it can be used in many different ways.. I think it will put FogCreek in front of more developers and managers and get them into more spaces like GTD (rememberthemilk.com) areas...
Kinda ironic after reading Spolsky's article "The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!" :D
Internal server error on the confirm link, and my full name has an "é" in it, so I guess it's the same issue. Also, log in with Google Account does nothing / doesn't work either.
Err... what? It's easier to shoot yourself in the foot, sure. But I haven't had unicode issues since I learned how to handle things properly years ago, when I was first getting into development.
Off topic: Proper, consistent and through UTF-8 support is one of the various reasons I'm starting to prefer Go to Pytho. It is sad that Python 3 didn't take the opportunity to properly fix this.
I thought Python3 sort-of did fix it, by forcing strings to be (abstractly) unicode and force you to explicit convert them to bytes with whichever codec you want (e.g. utf-8) when you need to.
Originally posted this on the announcement blogpost, but my comment is still awaiting moderation, and joel is posting here. :-)
Several major +1s:
1) Use of Google login, with ability to set a password to log in without that. I LOVE this, and it fits with what Joel (and Jeff Atwood) have been proselytizing for a while about the use of OpenID.
2) Awesome, responsive UI.
Also a few -1s:
1) No indication about pricing plans. Is this going to cost money one day? EDIT: I see now that you mention in the blog post that it’s free. And the site says "Creating an account is free and easy", but you know how often sites say that but mean "creating an account is easy, but to use our software in any meaningful way you’ll have to pay."
2) I had a problem when I created a new board. The UI took a while to respond, during which time I got confused, created another new board with the same name, and ended up with two new boards with the same name.
Suggestion: The menu that opens when you click the arrow in the corner of a card should open with right-click as well. This is how assembla’s card board works, and I like it that way.
Suggestion: Labels should take one click, rather than two. On the menu row for labels just have six colored squares to click on. Maybe that won’t work so well for smartphone users, but for a desktop, I’d rather save the click.
Not directly related to this string of comments, but as far as providing feedback... How can I do this? I don't see an obvious place to provide feedback.
Markdown code blocks don't seem to be working. (You can't put 4 spaces before code, but backticks work)
#2 occurred 15 minutes ago. Not necessarily related to that issue, but possibly another quirk.
Related: Is there a way to delete boards, or just to close them? I now have two junk boards (the original duplicate, and an empty board I just created to test) and I can imagine that list can get full pretty quick.
It definitely has all the trappings of a Fog Creek app:
- The overall UI: built by programmers who dabble in design
- A seven(!) minute video about how _simple_ it is to use. It's easy! Instead of just raising your hand, you mark a light on the corresponding tote board, which informs your manager that you need more information.
I hate to be contrarian, but there are many other apps out there that solve this problem much more succinctly. I'm not sure who thought a solution like this is needed.
But in typical usenet style, I will engage each of your points directly and anxiously await your rebuttal.
- "The UI stinks."
I think this is the number one thing we've gotten praise for. It's hard to please everyone, but overwhelmingly people have told us the opposite of what you are saying.
- "The video is too long / The app isn't simple enough."
I sort of agree with this. The video is kind of long for an app this simple. I got a bit bored in the middle (and I'm the Fog Creek co-founder). But if you wait until the Jello line at the end, it's totally worth it!
- "People already solved this problem."
Right and wrong. I think there are a few other solutions which are pretty close to this idea (some of them are even posted here in the comments, and we've since discovered more as people tweeted them to us) but I think we'll get more traction because we aren't building a software dev tool. We're trying to build a horizontal tool, and our future releases will reflect that focus. The majority of the existing tools are aimed at developers.
- "Joel pronounces lemur incorrectly."
I'm pretty sure he did that just to annoy you. It apparently succeeded.
No rebuttal here! Thanks very much for the clarification.
If you're getting praise for the UI, go for it. I think my biggest issue is the relatively narrow nature of the design. The cards, the columns, etc. Example: http://cl.ly/1i01452a3l0V1X3K3N1f -- line breaks like that are difficult on the eyeballs.
Little things, like using color-coded labels for organization sound good on paper, but are very tedious in practice. Lots of calendars use this with good intentions, but you'll find yourself asking, "is that a grey, or taupe task?"
Walking the user through the on-boarding process would help immensely in describing the product. Right now, there's a bit of information overload with everything being visible on the screen. It'd be neat to see an approach where you're presented with nothing, and are told to "create a list. Great! Now add yourself to this list. Wonderful! Lists are used for X, Y, Z"
I do apologize for being so gruff in my original comment. If you're getting great feedback on Trello, there's obviously something there. Good luck with it!
As someone who's used FogBugz and Kiln for some time, and has just played around with Trello, I'd say this:
FogBugz definitely suffers from what you describe. Despite what Fog Creek might say about it being easy, it feels clunky and indirect. And slow -- it's 2011, yet everything that manipulates the case table (arguably the most important part of the app) seems to require a full page refresh. It definitely suffers from the "you must mold your brain into our mindset to use" mentality. Furthermore, it's languishing -- most of the updates release to the On Demand version in the past year feel like small tweaks, many related to Kiln. Where's the leadership?
Kiln, on the other hand, is excellent. No problems whatsoever. Does what it's supposed to and does it well. No friction.
Now, Trello. I've only used it for a few hours, but already I look forward to using it for as many projects as I can. Simply amazing and simple and intuitive, all at the same time. It has negative friction -- practically invites me to put everything I have on it. I don't know what Fog Creek's doing in regards to product design, but it's working.
(And granted, most of those FogBugz problems could simply be because of a hefty anchor of legacy code. Maybe FC has simply moved on entirely)
Stay tuned for some significant changes to address much of what you've said about FogBugz here. Definitely not moved on. Just going to show it all to you at once.
I would say change the colors. They're as fresh as Windows software from the 90's.
The typography is rather ugly and makes the UI look stuffed (so maybe tuning the font, font-weights and font-sizes would help)
Overall the UI looks stuffed and heavy and colors make it seem depressing. The actual task something like two words, but about half of the space is occupied with heavy indicators of comments or other UI elements, so you tend to focus on the UI and not the actual information.
I guess the UI works in the basic level, but it's not that well or tastefully executed, and definitely looks like coder design.
http://agilezen.com/ is very close, and actually has a bunch of other useful features too. Although I've got to say that trello is a lot slicker than agilezen.
"Holy shit. This is EXACTLY what I was envisioning. This is freaking AWESOME. Since the data is stored outside of [our company], security might have a conniption fit if they found out that we were using this for managing internal project data.
I'm certain a tool like this could be highly useful to many other teams..."
The Trello site is written in plain English - it's easy for anyone to think of ways they could use Trello. The Walboardr site is full of 'buzzwords' that make it attractive to a segment of the market ('swimlanes', 'burn-up charts', 'points per status', 'backlog', 'iteration' etc.). Trello looks like it's trying to be a general purpose tool, Wallboardr does not.
Note that this isn't a value judgement - if it's _intended_ to be a niche product that's really appealing to one market segment (so they'd always choose it over e.g. Trello because of its specialization), that could be a good thing.
I initially wanted Wallboardr to be a fairly free-form board that wasn't specific to any industry, but as it developed I thought that approach was lacking in power for my primary audience which was definitely developers.
But it's very interesting to see what Joel and the guys/girls have done because I think they are hitting a sweet spot between power, simplicity and industry-neutrality.
If you can't customize columns, you've already lost :) But I really like the way you explicitly mention Kanban and Scrum, which Trello doesn't mention at all.
Thanks! Well customising columns is the obvious next step and is on the top of the list, so that won't be far away.
To be fair they also have comments and todo lists per card, like Pivotal, so Wallboardr is a bit behind at the moment. I've been wondering how essential sub-card todo lists are for a while, I've found myself wanting them, but MVP ideals have avoided so far.
Only found out because I wanted to view the source and it failed in Chrome so tried IE (turned out view source failed because of the load on the server atm).
Joel, btw, the favicon's missing, it's explicitly referenced in the source but returns a 404.
EDIT: Forgot to say like the look of it, good job.
Thanks for the nice words! I'm not sure about IE8, but, this being a new product, I specifically told the team not to worry about any non-current browsers. By the time this product hits its stride, IE14 will be shipping.
This looks really cool. I'm excited to try it out, individually and with others.
Even better that you can immediately sign up and give it a try. As opposed to Asana, which has similar intentions, but is doing a private beta with larger companies only. Yes, Trello is much easier to get excited about.
EDIT
Some feedback: so far I really like it. It's intuitive. And it has some nice features out of the box that are lacking in other products such as assignment and voting. Assignment allows central authorities to exist, as well as hand-offs between people. Voting is awesome. I immediately see two uses for this: democratizing, and to allow collaborators to vote on what they want to do. The latter being something I've always personally wanted in a collaboration tool. Combining that with a central authority can be very powerful by allowing people to voice their interests, yet keep the project moving and avoid conflict.
As for the initial reaction: some of the views are pretty intimidating. Even thought there's only three lists at the start, it is still a little much to begin with. A lot to take in. If you ever implement a "minimize list" feature, that could easily reduce the noise for a beginner while still allowing you to explore all the features when you're ready to.
The edit screen has a lot happening in it too, but I think that's less of a problem since it's only visible when actually editing cards and you'll soon be using everything in there.
Also, I'm sure it's already in the works, but keyboard navigation will be huge for upcoming power users. Right now Vimium provides that for me, but it'd be great to have it built in.
It looks very useful! I'm already having fun with it and I'll try it out on an upcoming collaboration. Is there going to be a UserVoice (or similar) site anywhere for additional feedback? I'll happily leave this and other feedback there.
The first thing I looked for when I picked up a card was a "recycle bin". Also, I cannot delete an organization I created.
I realize this is developing, but I would like to request some buttons for important actions. Some choices are not accessible enough, and thus require too many clicks to perform.
It would be nice to simply minimize the board and see a list of other boards I can open. For more complicated projects, as the user above noted, minimizing lists would be a helpful feature.
I would also like to be able to edit a card's text from the main screen without having to open it up and then click on the title again (e.g. to correct a mistake).
I was really excited about their initiative when I first heard of it. My day job at the time really needed it, or something like it. But it was private, and PM wouldn't jump on anything without seeing it in action, so that had no traction.
The idea of "flipping" the card over so you can see lots of details fixes my main beef with Pivotal Tracker: everything is so tiny (seriously, attach a screen shot and try to look at it) and you only get a certain amount of space for comments.
This looks like you could have a real discussion on the back of the card.
"vote" is a little weird... Everything about this is pretty intuitive except for that. It's only mentioned once on the summary/info page and in the welcome board after signup there should be a card that explains what voting is supposed to be, exactly. (besides just the ubiquitous like equivalent)
I get the impression this started out with "voting" more prominent than it is now.
But otherwise, this is really pretty neat. One of the first hosted solutions I've been interested in since I got tired of basecamp.
Voting is actually a kind of "test of concept" of the idea of Trello Plugins which enhance Trello boards in interesting ways that not everyone necessarily uses
Edit You CAN disable voting by going to Preferences under the main board menu.
(Disregard the following.)
Currently, no, because it's kind of a half plugin.
You can ignore it, though, nothing really changes if someone happens to vote on something. But internally, we've already found it quite useful for things like helping decide a name of a new project and voting on features.
Its cool that you can have a public board with public comments. But it would be even cooler to have a finer grained control over who can do what. i.e. have a board with some of the lists public read-only, others private, others read-write so anyone can edit, and public append-only, for things like bug reports etc.
since there is not feedback link I could find, I'll keep it here ;) :
- color labels are nice, but real text tags would be good too. lists are not enough. having a 'bugs' list doesn't help as soon as the bug moves to 'under work'
My first question is where's the API? This is something I plan to use but I will want some way to export the information to non-users, dead trees, etc. (I'd need it anyway for private boards, The boss wants it in excel etc.)
I'd also like to question the wisdom of closing the 'public' content off from non-users. Choosing 'Public' will make the board visible to all Trello users. Very closed-web, I don't use Orkut anymore and I wouldn't be on github if it wasn't so visible and ubiquitous. Not a complaint, just food for thought.
We were in private beta until today, and changing that wording got missed. Public boards are, in fact, visible to anyone, regardless of whether they're logged in.
I don't like this, it's telling me to go upgrade my browser as punishment for having the wrong user agent. When I'm trying to have privacy.
My agent:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.7) Gecko/20091221 Firefox/3.5.7
I'm using 6.0.2 browser, however.
It's a pretty common user agent: https://panopticlick.eff.org/index.php Eff does not provide the most common user agents list so I have no idea what's latest user agent to use. Many web statistic collectors also don't mention user agents.
This product was designed to do a really good job helping people that have to get things done. Like any other software project, they must have had thousands of issues and limited resources to fix them, so they had to triage. With all due respect, the kinds of users that spoof their user agent to protect their privacy is probably fairly low on their priority list.
"Thousands of issues and limited resources to fix them"? They seem to have the resources to make it by design to block such user agents rather than advising users to upgrade.
Firefox 5.0 is the most common nowadays; XP is also still the leader; So, the intersection between the two is probably high. According to UserAgentString, there's only one string (well, two variations):
Mozilla/5.0 (Windows NT 5.1; U; rv:5.0) Gecko/20100101 Firefox/5.0
Mozilla/5.0 (U; Windows NT 5.1; rv:5.0) Gecko/20100101 Firefox/5.0
By the way, have you considered using this[1]? Seems the best option to prevent tracking.
From what I can see it's a lot simpler than Basecamp - it seems to be great for a single view if lots of small things rather than managing the detail of larger things (if that makes sense).
This looks really great. I've been wanting to write something like this for use in our small-ish team, but there is probably really no point if we have this.
I would love to see more e-mail integration: someone mentioned mailing in to it to create a card, which could be pretty cool, but probably just as importantly I'd like for it to have options to send us e-mail when cards change. Some people like getting notifications pushed to them via e-mail versus having to check the site.
This is really cool. I like it a lot. But I think "organize everything" is a bit of a hyperbolic catchphrase. It seems to be very task-oriented. It doesn't feel like something you can just throw random ideas, say for a screenplay, into to organize them.
Still, as long as it stays free or even inexpensive it might be my goto tool for task management. I'll need to try it. I like that it works on my iPad without much fuss.
Wow, this is seriously great. The one thing that stood out to me (from the blog post) was the wall of 42" Plasmas in the office just for displaying Trello. While obviously not everyone can afford a wall of TVs it really would be nice to have a way that everyone could always see who should be working on / is responsible for what.
I imagine a board for "existing modules / features" That has the current "responsible" person for that item.. and a "new feature" board that has an easy way for people to see who is currently implementing said new feature. Though that would overlap somewhat with fogbugz (search for the task and see who it is assigned to) the board would have the advantage of being more high level and still being easily visible after a task is complete.
Joel (or FogCreek persons), I'm using fogbugz / kiln at home and for my side projects is there some plan to provide integration with those existing projects? Magically linking based on case numbers? Updatet he responsible person on a card based on who is currently assigned some case number?
I'm starting to use the service now and it immediately occurs to me that an excellent feature would be the ability to forward an email and have it create a board / card for me a la fogbugz.
Very slick UI, the use cases are virtually endless.
How about email integration where you can forward emails to different lists? Is there an API that we can integrate with Mailgun?
Use case I had in mind is for a sales funnel (or any funnel) where I can bcc the list corresponding with the stage in the funnel as I am corresponding with a lead and have the email move through the lists accordingly.
It's funny how the tiniest moments can really give you a glimpse of the type of person someone is.
For instance, in that demo video, at the end when Angella Kim makes the reference to Jello (in the heat of the moment) was one of those moments that makes me want to just give her a hug and put her in my pocket.
Also, this product looks good. I am wondering though, what will this cost and how will I be charged.
I hate that it just says free right now...with no indication about how this will be maintained.
I would hate to start using this, just to see it disappear in a few months - because it was free only. I know that if they are wildly successful and it starts racking up big bills they can charge for it, but I want to know how will that affect me. I trust Joel to do what's right by early users, but this is a concern I have with new stuff that I don't see a sustainable path.
I will probably still create an account, but not knowing whether this can be around, or I will be charged in 6 months after I am addicted is a bit annoying.
According to the blog post (http://blog.trello.com/launch/), "It’s free. (We might charge something for premium features in the future). You can make one board or 100."
I'm using this already. I'm managing 3 clients right now and I already can see the benefits. I can see everything - so will my brother (co-hacker) when he accepts the invite! I'm using this as a to-do list.
I'm still going to use this with caution. I do not want to rely heavily on something that I could not afford in the future.
Hopefully, the pricing/freemium will not make me back-out.
My first impression is that Trello isn't a (good) product (yet) but it will certainly leverage on Joel's marketing machine.
Observations:
The organization information should not be public by default. I haven't found a way to delete an organization (is it there?).
The interface is unusable on iPad/mobile.
The card pop-up window is hard to use when there's some actual information attached to it. The bird's eye view is confusing and offers little information.
Activity log grows fast with information I would not need: voting events, add/remove members, etc. Make two activity logs, one with useful information and one with tracking (investigation) information.
In Opera and Chrome the red connection establishing notice appears all the time.
The in/out/public permissions are easy to use, but users may actually need more granularity.
Most exciting thing about it is that it's written in CoffeeScript & Node.js.
I really do have a feeling the combination of the two is going to be the Ruby On Rails of this generation of the web. And the shift from server-side development to client-side is going to be a huge one.
Most of the kanban boards out there worth their salt cost some serious pennies if scaled to a large team!
plus... as far as I know Trello doesn't have the functionality for WIP Limits
This board also integrates nicely with GoogleApps, if it has integration with tfs (not sure if it does) then it would stand a good chance of owning a piece of the PM scene in .net land
Looks pretty awesome in my 5 min trial. Back of the card is very well done.
This will could go way beyond the software crowd to a general organization app..I just hope this thing scales well and they clarify their upload limits and such.
Edit 1: I tried their iPhone app and it's very far from their web interface - took me 6-7 clicks just to get to a checklist for one of the items. It's commendable that they have a app on launch though so I am sure they will work out the app interface with time - currently it's an order of magnitude less usable than their webapp.
One of the devs whipped the iPhone app up in essentially 2 weeks after some of our earlier plans with a contractor fell through. It was literally meant to be there on launch read-only. We know it's got a ways to go, but given the time constraints it was the best we could do (IMHO it was better than anything I thought we'd be able to do... Justin really cranked on it)
- I do the exact same thing, but using post-its. Here is a photo of one of my early methods: http://i.imgur.com/hEVtT.jpg. Later I started using post-its for the tasks, so that they could be resorted.
- The most important productivity reasons that I noticed AFTER using this method were:
1) Limits to a few projects on my screen at at time
2) Can assign priorities by drag-drop (or unstick and paste)
3) Can see projects and tasks in one look
- The point is, that if I had time to develop an app to automate my manual workflow, Trello would have been the exact type of app I would have made, verifying that software development project managers are also going this route.
How is this going to be monetised? I want to use it but just want to make sure we don't get too excited/committed if it's going to cost a lot of money!
I tried to solve this using WhoIsWorkingOnWhat.com (hn submission: http://news.ycombinator.com/item?id=1979671). It didn't get much traction. I was obviously a no match to Joel. I am thinking of open sourcing the app when I get some time to clean it up.
The home page clearly explain what the software does, the signup process is really simple and straightforward, and it's free.
Once you get inside, you have a fake board which is there to help you get started.. It took me 2 minutes to try and understand how everything worked. No magic, no complicated features.. really simple and intuitive.
I love the "reverse of a card" concept, the small animations when you drag a card, how you can easily "add people" to cards, how an avatar is automatically generated for you (With your first letter and a small icon), and more importantly, how there're just a few well-done features instead of a thousand of useless and over complicated stuff.
Furthurmore, it seems that the app introduce the concept of plugins where anyone could potentially incorporate only the features they want.
Does it look a bit more complicated and crowded version of Co-human? http://www.cohuman.com/
I'm on the fence, I kinda like both and love neither. Especially so far none of these organization tools helped me to be more efficient. I guess I'm holding them wrong.
It looks intriguing to me, and the page does give the impression that it's nicely designed, but I have no idea what it does. Something to do with tasks, and project management, and it "works on mobile too". Maybe I'm expected to know already kanban, what it brings to project management, and how it works?
Watching the Trello video, and browsing around the http://trello.com/ landing page, I immediately have ideas about how I could be using Trello - that's not the case on http://www.kanbanpad.com/.
We try to lower the barrier to entry so much so, that if you just type your e-mail and hit "let's go" you can already start playing with it, instead of having to watch a video.
However we are putting together a nicely animated video that should explain it with a nice elevator pitch.
Voting and commenting on public boards can be allowed for everyone, restricted to members, or turned off. We chose to leave it on because we like the aspect of simple, quick feedback on where things are going.
Played around for a few minutes.. I didn't see a quick way to filter the view to only the non-archived cards that you (or any other single person) are a member of. If you start to build up tons of boards/cards, is there a way to quickly filter the view like this?
In general, I think more work is needed if the numbers of everything (members, cards, etc) all scale to large numbers. I love the interface and would actually like to have this internally, but I worry that things would get out of hand quickly. I think board hierarchy (nest boards within boards) could go a ways towards making it better, but I'd have to play with it. They should try scaling up to large numbers of boards/cards (1000?) to which you are a participant and see if any of the concepts need tweaking.
This application uses other third-party javascript components distributed under appropriate licenses. For more information, see the following files at http://trello.com/js/lib/
Joel - if you're still reading this thread (it's getting long) - I've been looking for a tool like this for a long time, but the lynchpin for me is the ability to track work hours so I can bill against it. I constantly run into tools that either let me track time or projects but rarely both in a way that works. This web app works well, but without time management, I'm forced to track the very same projects in another system for the sole purpose of billing. Oh what can we do here? At the very least, is there an API where we can add time tracking and billing to this?
Since reading the news about the release, I thought I'd try it out on a wee project I'm working on. So today, I tried to remember the name, and guess what? I did. Good job with that. I went to Trello.com, and found out I could just click the login with Google button. Yes! No forms or anything. I'm happy. I'm now logged in and it's taken me a total of 3 minutes to get fairly comfortable and start work.
Thank you so much for this service. I hope free accounts are grandfathered in :)
p.s - Please make it so that I can invite more than one person per click.
It looks so similar to KanbanTool [http://kanbantool.com] (intuitive UI, board, easy drag&drop, comments etc.) - but it is much less customisable and powerful.
I'm just wondering if Trello is a finished product or do you guys planning to implement real-time updates, notifications, history, more customisation, priorities and any features like KanbanTool has at the moment?
And is there enough place on Kanban market for another tool?
So I guess with the top navigation bar of Fogcreek.com having lots of available space, we can expect to see a lot of new offerings from Joel and company?
Can someone confirm a bug or let me know if it's just my browser? (Firefox/XP)
View back of a card in the first list. Choose Move.... Try to move it to a specific position (2nd, for example) on another list (in my case, the third/final list). For me, it just goes to the top of the list.
Edit: Tried to use this move method (rather than dragging) to shift something from List 3 to Position 3 on List 1. It shifted to Position 2 instead.
Mostly a nitpick, but it'd be nice if the home page had a little <noscript> text for those of us who browse safely and are wondering what it is at a glance or why we should care enough (besides the Spolsky reference) to enable scripting.
Anyway, it's pretty slick with the UI. I'm going to check it out for tasking myself, and if I like it see how it works for a school project with others.
What kind of offline support do you foresee? I haven't tried using the iOS app yet...but I'm thinking of the use case where I've cached the current state of the board and want to check off/add things to the board while I'm on the subway, and have it sync automatically when I get back on. Possible, or are there too many moving parts for that to be implemented easily?
I couldn't log in with my Google Account (approval with G worked fine though).
Anyway, I created a regular account and my initial impression is very positive. I feel a little disoriented and the "See all boards" could be a lot better (make the boards and their relationships easier to figure out visually). Will use this for a while to see if it will grow on me.
Amazing! I'm starting to use it today, most todo/list/project apps suck, this looks really good -- so far =:)
The only thing I don't like is that there is instantly a member added to the first board "Trello" - which makes me wonder if somebody is looking at everything I post on the board -- and makes it seem less private/thrust worthy.
I was pretty skeptical after my hate-hate affair with FogBugz but Trello's design, marketing and OOBE are far and away the best in the category. I need to use it more but it just doesn't have whatever Pivotal has that scares away non-devs at first glance. Joel, this is definitely my favorite Fog Creek product yet.
The only thing I don't like about this is that my company won't own the data if we use it. By own the data, I mean, be responsible for it entirely, without any outside entity having anything to do with it, whatsoever.
Other than that, looks awesome. I hope there is a standalone installable version of this somehow, some day ..
Nearly freaked out when I saw this as it's very similar to something I'm working on but, on closer inspection it's just a surface design similarity.
Glad to see positive reaction anyway, sometimes seeing a product that's like to the one your working come out isn't a bad thing, it just tells you your on the right track
Getting an internal server error when trying to signup (after filling out the signup form and clicking the "Create New Account" button on the signup) ... http://cl.ly/020I2N0Q3d143L2f040a
Any plans for adding multiple lists per column (two at 50% height, etc)? That would be the only thing I can do on our current Kanban board that I couldn't do on Trello.
I really like the app though, it's like a distilled, get out of the way version of Jira.
Maybe - I think that the border around the lists is clear enough separation. I agree about the horizontal scrolling, that is why we have multiple queues per column on the whiteboard. If you have lists with only 2-3 items in them at a time (think WIP limit) then the bottom half of the screen is empty.
BTW, I realize now this could be read sarcastically. I didn't meant it that way at all. I know from the blog post it took 9 months, but I'm also curious the composition of the team that built it.
There were two full-time devs, two designer-devs (they write code, for example the iPhone app, but also spent lot of time on UI, usability, and design), and two interns for 12 weeks over the summer.
Just signed up - great so far. One request: make the back of the card resizeable to variable width, or have a preference to make it fit-to larger area within window. Wrapped c++ stubs look even uglier than regular c++!
Love it! It's basically a todo checklist app with a completely rethought interface...and general enough you could use it for everything from software development to portfolio management to sales pipeline review.
Great! This is exactly what I was looking for a long time for our content generation and feature implementation workflow. Idea board => Doing board => Done board. Thanks, Joel!
When signing up, the verification email does not content a link if viewed in text mode. My email client is configured to show the text portion of the email when it's available.
Have used it for a week and absolutely love it. It would be even better if they could add offline access so I didn't have to worry about having internet access all the time.
I have a number of small business clients who have been looking for something exactly like this. I'm going to be telling each of them about it the next time I see them.
Feature request: reorder boards. I'd like to use this to manage our software project, and I can imagine a ltr task progression, but then boards have to be features.
Hey, cool, a nice looking Kanban board! (The ones I've seen to date have either been pretty ugly, or too many features for what I want. This one looks just right)
If you're really interested in introducing this into a working business, and have spent days researching the options a 7 minute summary (advert) like this is a gift.
We have the hosted versions of both and I've heard more than once "I can't believe they(fogcreek) doesn't do something in project management because most of the current solutions suck."
Minimally for now - a FogBugz link in a comment will show a little kiwi icon on the front of the card, which you can click to follow that link. The planned API will likely lead to a plugin.
Man - I would kill for this. We use fogbugz / kiln right now, and integration into these boards for cases / users / pushes would be the icing on the visibility cake.
is it just me or does the background music make this demo seem like a movie preview?
seems like Joel is about to meet his long lost brother, who at first, completely ruins his life, but in the end makes him learn a lot more about himself!
Agreed. I have a strong suspicion this will cause me to throw my tiddlyspot out the window. Per-board privacy security allows me to have my "Feature 1" board, "feature 2" board, "Bills to pay" board, etc mixing business with pleasure in a way that tiddlyspot doesn't allow.
Since everybody seems so enthusiastic about it, I thought I would also leave my not so enthusiastic opinion for the sake of broad feedback.
If i should be honest I didn't like it so much. It felt too cluttered, the interface has way too many visual elements for my brain to process in efficient time. It's also missing more obvious visual indicators such as color or shapes. The list look all the same, they don't even have different icons identifying them, only the the name. That will do it but it's not the ultimate visual indicator.
As a person that barely uses the mouse, I don't find this so practical, it's click after click after click, but I guess that problem affects almost every web application out there.
I might be too focused simple/minimal things, this tries to lay information in a rather complex data structure, which in practice it means a lot of mental exercise before you get the info. I believe many like it, not me, give me a search box and a list of results every day.
In the blog post, you mention that these teams are adopting the Lean Startup philosophy of ship early and ship often. Does that mean that you've softened on the "never ever write any code without a spec" dogma from "the Joel Test"?
I've personally found that when you're doing tight iterations and continuous deployment, writing old-school spec documents feels, well, old-school.
If I can go straight from human conversation w/ whiteboard sketches to working, tested code running on production servers without creating intermediate written documents, I think I'm winning.