At least they tell you what's allowed. My favorite is when I put in a password that's rejected with an ambiguous message. Or, even better, when my bank accepted my password but it actually didn't handle whatever punctuation was in it correctly, so that changing my password succeeded but I couldn't log in at all (I was using KeePass, so it wasn't typos).
Had this happen with a credit card company. Registered for online access, generated pass with 1password, no errors, "Very strong" indicator, save (again, no errors) and was locked out.
Was a lot of fun talking to the customer service rep who insisted I needed to be using IE. That I had to register. That I "must be doing something wrong". That I am not typing in the correct password. That I'm not technically capable.
Turned out to be a length restriction. It just cut off the last n characters of the password I chose. Good times.
Yes! My bank did something like this - they kept rejecting my password as too long without telling me how long it was supposed to be!!
Turns out it was ten. Ten characters protecting my sensitive personal banking information. Upon e-mailing, they said they're going to be bumping it to 20.
My developer environment at work has a password that is synced across multiple services.
Ran into a problem a few months ago where I changed my password successfully on the front-end, but one-or-many backend syncing operations mangled the new password by dropping the last n characters on the floor. So when I logged into the front end, it would look like everything was fine until I tried to perform some kind of operation. At which point it promptly threw up all over itself.
Left me in a completely non-working state for a few days. Didn't help that I'm basically the only admin for said system.
Non-responsive superiors are a great excuse in business. Stuff can get "stuck" for quite a while if you have enough bureaucratic obfuscation to justify it. When the buck starts and stops in the same place you have nowhere to turn.
At my bank I'm restricted to a 5 character password. When asked if they think that would be secure enough I was told that an attacker would also need the login name and that should be kept secret as well (default login name is account number or FirstnameLastname and I doubt many users will change that)
This happened to me on Mint.com. I use a password manger as well and I generated a 32 character password. It saved successfully, but once I'd log in, it'd kick me back letting me know that my password was incorrect. After about 15 minutes of confusion and multiple password resets, I figured out that passwords can only be 16 characters in length (or, that worked at least), and anything longer than that amount was just being ignored. There was no explanation about this at the time.
My concern here is that if there is a restriction on what passwords can contain, how are they being stored?
It seems concerning that between entering and hash/(b/s)crypt-ing passwords there would be any step which required these limitations. A regex or similar validator on strength not equipped or written to handle other characters? Either way... really?
This drives me nuts. Every time I encounter a website that was setup with constraints it drives me nuts. Could probably find a few rants on my Twitter feed without much effort. I actually just sent an email to my new utility provider complaining about this over the weekend.
I use mind maps all the time. They were instrumental in helping me pass GCSEs, A Levels, UnderGrad and parts of my PhD. Before I read the book (Tony Buzan, Use Your Head) I did OK in exams and school but nothing to write home about. With mind maps I went from a C to an A across the board.
Whats most amazing about them, for me, is that even after 7 years I can still remember good chunks of my A Level Physics and Electronics mind maps.
The key is to read the book and use them properly, most people use them as a glorified spider diagram (like this software). A good MM stays in your head for years and it's like feeling you way to the information rather than just brute forcing it from memory.
I highly recommend reading the book.
I have 2 MMs on the walls next to me. One for how we are going to get enough traffic to our site next winter and another detailing what we learnt last winter in general in the business.
- Loose sentences/notes, which I group into 'clouds' of related topics (without a mindmap's edges). Scapple [0] on OSX is good for this (and spider diagrams), and Microsoft OneNote.
- Spider diagrams - I find mentally these much easier to create when listening to a presentation or in a meeting. The straight lines help for some reason
- Mind maps. My weakest skill, but as someone else commented they stay in the mind better than anything else does, and I can remember some from 3+ years ago.
It's difficult to describe what I call "properly" without doing it a diservice but the original book is called "Use Your Head" by Tony Buzan and I would highly, highly recommend it. The first half doesn't really touch on mind mapping and is more on how memory works but that sets up for the mind mapping section and is just as (if not more) valuable.
It's an old book, I read my Mum's copy when I was 14 and it was hers from uni but it is great.
I find that students split two thirds to one third on this. Two thirds prefer 'linear' lists of topics and notes, the one third like the mind-map approach. This is 16 year olds and older studying basic maths. So I have both available!
Personally I don't like the 'strict' mind map format, but do like a two dimensional spatial arrangement of information. (Dia rather than freemind for the Linux desktop users out there. But preferably a pencil and paper).
I find these mind mapping programs annoying and frustrating to use. The thing is, these are not 'maps' in so much as they are a link of concepts. On a map, the placement is the most important thing. In these softwares, your three element list just became a three branch tree, or a spiderweb. Does the word on the left mean something different because it's on the left? Why not put everything on the right of the root since that's how people read anyway (if you read from left to right). Why does my eye have to jump around the whole screen just to read a list?
> Personally I don't like the 'strict' mind map format, but do like a two dimensional spatial arrangement of information.
Completely agree with you. I haven't found any software to match using pen and paper either, which is sad as when I run out of space around a topic, a force-directed algorithm to create more would be very nice :)
As part of fiddling with a different way of displaying dependency information, I wrote a series of blog posts talking about how people don't think hierarchically.
I never really understood the appeal - I don't see that gives anything beyond what a simple tree control based UI can deliver. Perhaps it's a matter of taste.
Flashback. I worked on a failed mindmap startup with a web-based tool just like this (it was 1999 so of course it was Flash!).
As the other commenter said, there's a split in people who learn well with mindmaps and others like me who prefer lists.
The startup I worked on: Mindwarp Pavillion in Dundee got licensing rights to stacks of textbooks. They created loads of mindmaps based on study books and sold access to students for a few bucks a month.
Even (better?) the mindmaps were a quiz where the student had to answer the next node. There were studies showing that using this method they retained a lot more information.
I was told that the whole mindmap concept was protected somehow (patent?) which they paid a license fee to use. They also got the endorsement of its creator.
£30m valuation when the local authority invested. They lasted a year then died, leaving my last invoice unpaid.
They failed because:
- Students didn't pay
- They could do it on pen and paper for free, while they're learning
- The product wasn't driven by a real customer need: 12 months of dev on super-whizzy software without getting a MVP in students hands
"They could do it on pen and paper for free, while they're learning"
And that active engagement with the text is the important thing! The student making their own mind map/flash cards/linear notes/annotations on past papers. Whatever.
I like the node quiz idea though, and shall use that on the whiteboard one day (UK based maths teacher).
Indeed, the most valuable part of flashcards for me has always been distilling a semester into a set of questions that are suitable for the format. Then the flashcard is a marker into a larger corpus of information that is in my head, instead of the sum total of information I have on a topic.
Is there a demo video available? I think you should make one if you don't. Most people I've talked to having videos that outline basic way of using a product give extra 10-15% bump in signups.
I am for one at this point is a bit tired and would like to watch a happy video about your product so I don't have to confront my doubts and insecurities while learning subtask.
Maybe I am wrong but thats the first thing I looked for on your site.
I like these services, but it's like trello. If I have personal tasks to do, those personal tasks will often include sensitive data. My grandmothers bank number, my medical card number, the phone number of a friend. I don't trust these services with that kind of data. Having an online solution is not a solution. Not without zero knowledge. Which they could totally do, but hell, that's hard.
That's where local is beneficial. I've got a 'the hit list' install locally on my mac that stores just about everything & a bunch of flashcards related to each task. There was a learning curve but it's immensely helped me stay organized.
Why? If it is a feature that will encourage people to upgrade, and/or a feature that only concerns the kind of people who buy "Enterprise" plans, then why not use it as a cost differentiator among the plans?
I can see how SSL is important, but the 2 lower plans are free and a few bucks a month. I don't see a problem with reserving this to the bigger plans.
Pay-for-security is a bad path to go down. Compare pay-for-ssl to using md5 on the lower tier for password encryption, while the upper tiers get something like bcrypt, or you only get a salted password if you pay extra.
It seems pretty absurd to require a payment for security, especially when you're implementing it for a subset of users. Its true that SSL is going to be more taxing on their servers, but the majority of the cost is going to be spent getting an engineer to implement it, rather than the actual operational costs.
SSL concerns the kind of people Subtask is marketing its product to. Including me. It's a non-starter as-is. I was interested to sign up and try it out right until then.
Security is never a feature. The app lost all credibility at that point.
Normally, I'd completely agree that if it's a feature people want, then they can pay for it. But SSL isn't just about protecting the person's data, it's about preventing people from snooping their login credentials over unencrypted traffic (e.g. at a coffee shop). If they use the same login as their email or other accounts, then by excluding SSL from any tier, you're putting those users at risk, not just within your own app, but for their other accounts as well.
In an ideal world, if your app has the ability to login, it should have SSL. And I'm not trying to be a judgmental idealist either, just answering the "why" question after thinking it through. I'm certainly guilty of having a couple old apps out there I've not yet updated to use SSL. I think I may have to go do that now.
You're completely right and in fact all URLs where your login credentials are transfered (login, signup, change password) are guaranteed to be SSL regardless of which plan you're on.
If SSL is too expensive for the Basic plan, consider dropping it altogether, or increasing it to EUR 9.95. You can also slash the free plan in half (1 project, 5 MB, 50 tasks) to encourage people to move on to paid; one project is more than enough to decide if the tool is for you.
I can't help thinking the github model might work here: publicly viewable mindmaps are free and you pay to restrict access to specific signed in users. Privacy is a better selling point than security, and some publicly shared mindmaps will generate backlinks.
Things like login, signup, change password, ... always use SSL. In fact, currently everything else runs over SSL as well (even if you access the site over http you're redirected to https), it's just that in the higher tiers you are guaranteed that this will never change in the future.
Edit: Maybe I should add that I'm the developer of Subtask :)
Two notes:
- I think your Enterprise and Plus plans are too cheap.
- I'm glad nobody's complaining about the prices being in euros instead of dollars. As a European I sometimes worry about exchange rates affecting my income if I price my products in dollars.
- If you have selected a bunch of tasks, that's no use because you can't change them all at once (like assigning someone to an entire subtree, or setting a date)
- (WRONG there is a keyboard shortcut overview in the help) I have no idea what keyboard shortcuts exist and there doesn't seem to be a page that lists them and only sporadic hints on the elements themselves
- Links between tasks can make things very messy, but I'm unsure whether this actually gets problematic with more use
- For proper project management priorities are important and should be settable
- Tasks should maybe have optional weighting so that if you have 2 subtasks, marking one as done doesn't mean the task is 50% done
- Performance seems sub-optimal. While I don't experience problems during use, it is eating quite a lot of CPU power for me for things like moving around. Not a huge problem and I'm not sure this can be changed easily. Speculation: Maybe using a canvas renderer (2d or webgl if available) could prove faster. (http://jonobr1.github.io/two.js/ maybe?)
edit:
Some form of hover effect on interactive elements would help (like on the icons below a task)
That is the other sale that needs to be made. As an HN'er, I'll Wikipedia up an answer and Google around on it, but how am I going to sell this egg basket to my boss for our next project? Or rather, how are you going to help me sell it to Mr. PointyHead?
Looks awesome and I'm definitely going to be trying this out. My only initial concern is about how quickly I can input a long list of tasks. I don't see anything to add a sibling tasks; a button (with keyboard shortcut) for this would definitely speed that up.
Also, it would be nice if when editing a task, it could centre itself onscreen or display the text you're typing in a floating box. Otherwise you quickly find yourself typing off the edge of the browser window, having to stop typing and pan with the mouse, which is a minor inconvenience.
What are your plans for tablet support? I currently dump all my initial ideas into iThoughts and then use something else for managing an active project. I could see this replacing that workflow with one thing.
Thanks for your feedback! You can create sibling tasks using the Enter-key and subtasks with Tab.
Tablet: Depending on what tablet you have the web-version might work but we also plan to build native apps.
From my experience, it's almost impossible to divide a whole software project into a hierarchical view like a mind map.
Take for example a simple website. What will be your first level? Server-side and Client-side? What about pages? Do you repeat every pages on each side because you have code on both? OK so maybe the first level is all the pages then? Now, do you repeat client-side and server-side work for each pages?
One thing that is important to me in mind-mapping-type tools is good keyboard navigation. This one is pretty good, but I think going left then immediately right (-1+1=0) should end on the same item, not the top of the list of children of the left-hand item (and the same for left-left-right-right etc.)
Looks interesting. I've been using DropTask (https://www.droptask.com) recently which is a slightly different take on visual task management. Still not sure whether I prefer it to more traditional apps, but kudos to both companies for trying something different.
The odd quote placement here made me laugh, though, because it almost looks like you're trying to take part of what someone said and change it by adding more: http://i.imgur.com/Ea77zIr.jpg
Great tool, greatly increased the productivity of our group. The visualization gives you a proper overview of what's been done and what's still todo. Kudos.
looks slick!! congrats! Is it similiar to MindNode on the Mac? I use that a lot and have been on the hunt for a great web-alternative with no success so far!
It can be used as a mindmapping-tool, yes. However, it's actually more a project management tool than a mindmaping-tool, so some features like change the color of a branch are not available but you get all those PM features instead (like a calendar, comments, file attachments etc).
It's 2013. Why are there still constraints on what characters can be use for a password?
I would love to hear from the author why this is the case.