Ask HN: Review my webapp - Goalbot | Hacker News Reader

Ask HN: Review my webapp - Goalbot

goalbot.org

20 pointsmmorris35 comments

Really like this idea, it's a really nifty concept. On the design side of things I'm on the fence, I like the logo and serif font at the top, though I'm still unsure about the monospace caption, it's possible the entire header would look better in a monospace font, worth exploring.

From the design of the header, the buttons and heavy gradient don't really work for me, you start out very crisp and clean, but quickly clutter with excessive gradients and rollovers, and the central page layout feels very 'cramped' as well, possibly playing with the spacing between objects and the contrasts would help.

Honestly I think your design would greatly improve with a more liberal use of whitespace, possibly no borders for the content, no boxing things in? and I'm not sold on the rounded navigation either, but this is possibly just due to the images not really jiving with me.

Here's an example: http://thismat.com/images/goalbot.png

Obviously this all subjective to taste, but I thought I'd throw it in as food for thought...you might not even be ready to tweak on the design yet.

Wow! This is really helpful. I've often felt like the design was sort of cramped, as you say. Unfortunately, I'm only an aspiring designer so sometimes I can't figure out just how to fix my problems. I've also been a bit hesitant to play around with the design too much as I've been trying to get the functionality into place.

But you've inspired me. I'm going to play around with the spacing and borders this weekend. At this point I don't think I want to do away with the content borders entirely, but your graphic shows how much more open it would be. Hrmm... I'll have to ruminate on it.

I know the screenshot wasn't a super intensive edit or anything, but it was really nice of you to go through the trouble. Thanks so much!

Not a problem, for the record I just used firebug to hide the background images and took a screen shot ;)

Ha! You sly dog.

Hi, I triggered a 500 error somehow.. probably because of my javascript being disabled. You should add a small but very visible warning that js is required. Wondering how many hn-readers are using noscript. About the whole concept. I do not get it. At least I am not sure if I do. To be fair. I spent less then 90 seconds on your page. I would recommend that you produce some sort of workflow-demo. Wink (http://www.debugmode.com/wink/) does a nice job. I use the linux version to do the screencapturing and use the more advanced windows version to add audio. Do you have a business model in mind? Or is this just a fun project? Best wishes, have fun.

Yes, I'm still working on getting the non-javascript enabled functions working properly. At the moment the site does require JS, but it won't when it's "done", though the UI will be somewhat clunkier without the JS.

I've thought about adding a flowchart of some type, but in actuality there isn't really a sequential path through, other than having to plan your goal(s) before you can track and connect on them. I'll considerate further and see what I can come up with though.

Thanks for the feedback.

A tried a few things like: "go to law school", "develop a video game", "work on open source", none of which came up with any suggestions. I really like the idea, I think the database just needs some population before it becomes really useful.

Also it would be super cool if people could have public profiles where they post how they did things. You might be able to pay/interest some authors or famous people (like Neil Gaiman or Steven Fry, both of whom LOVE services like Twitter). Get them to write public how-tos on achieving your goals.

Absolutely, the site needs a lot more content before it's particularly useful. Those are all great goals, I'll definitely be doing some research and developing content in those areas.

I have a lot of thoughts on where the site is going, and how public profiles will work, but I hadn't particularly thought about the "famous how-tos" angle. Thats really interesting, I'll add it into the brainstorm soup!!

Thanks!

I'll be honest; dialogue boxes appearing everywhere put me off.

Thanks for the honesty!

I'm assuming you mean the pop up dialogues when you've started planning a new goal (and not the tool tips when you hover over certain items). This is really designed for first time users to help explain how the system works.

Do you think that level of explanation is unnecessary? Or is there a better way of adding that information?

no I think it is good to see it.

Just within a few seconds of getting to that screen I had about 5 boxes appear in front of me and it was a bit confusing.

Maybe just a reorganisation of where they appear?

EDIT: ok I went back to have another look. It's not the placement that put me off; it was the speed. It sounds crazy but I think the popups are too fast in appearing - I bet if you got them to fade in a little slower (2-3 seconds) then it would feel more natural.

Okay, I'll play with the timing and see what I can do. Thanks for taking the time to look again!

EDIT: I slowed it down a few seconds, but from the feedback so far I think I might need to rethink how the instructions work exactly...

I think if you could programmatically allow one pop up at a time, it would make a huge improvement.

That's a good point. I was thinking that I should just combine them into less popups, but it's probably best to let people have more instruction only if they choose to see it, which could be implemented as per your suggestion. This will definitely require some further testing.

Great suggestion, thank you.

Not to be nitpicky, but in startups every little bit helps, you should change loose weight to lose weight in the "start a business" goal. Loose refers to the opposite of tight. Lose refers to the act of removing something.

Other than the dialog boxes that pop up like crazy the site is interesting. One thing you might want to elaborate on is how the system uses the data you provide to fine tune the algorithms.

If you are crowdsourcing the data for goals how are you going to prevent malicious activity from corrupting the goal set?

You're absolutely right. I should fix that title.

The "loose weight" entry was from a beta tester and I haven't quite finished the 'similar goals' part of the database, which will connect "loose weight" to the correctly spelled goal "lose weight" (which is also in the database). That should be working soon, which will help to minimize accidental entries.

In regards to malicious entries, this is a tricky subject. I'm thinking of malicious entries in terms of spam-type and just plain junk entries, is that what you mean? To some degree, worthless entries will be weeded out because no one will select them, so they will fall to the bottom of the recommended goals list (more to the point, the opposite will happen - the good will rise to the top). There are also a number of other steps I'm taking, including implementing a flagging system that will help with manual moderation.

Thanks for taking the time to check out the site! I'm going to be making a number of changes based on the feedback I've received here. I'll report back once things are further along.

No problem. The site definitely has potential. It might be a good idea to add some kind of chronological aspect of the goal system into it too. By that I mean to have the goals sorted by chronology in addition to rank. Most goals have a linear progression that if the suggested goals are ranked in a manner that is consistent with the natural progression of attaining the goal it will help in make the system more usable. If you wanted to implement this system you could take a look at how people order the subgoals when the goals are added to the main goal. You could also keep stats on what goals are added in addition to the goal that someone just added and do a netflix style "We noticed that you added 'get married' as a subgoal to your main goal of 'live a full life' might we suggest 'find a date' as a correlated goal?" or something along those lines. Good luck on it and I look forward to seeing your progress.

I like the design, although the dialog boxes are somewhat annoying. I don't find it useful to me, but the concept is very interesting and original. Good luck with your webapp.

I've slowed the popup dialogs down a bit, but I might need to rethink how it works based on the feedback here.

Thank you for taking the time to check it out.

I'm curious whether you're planning to have Goalbot automatically learn from the collective successes of your users (example "people who skipped this step were twice as likely to fail at goal X"). Surely some goals are going to be personal, but I could imagine a lot of interesting insights arising from studying the paths of those who were successful.

Yes, the plan and track sections of Goalbot are based around 'learning' which subgoals are most effective in reaching the parent goal. This was actually part of the original inspiration behind the idea.

Also, I agree that there will be some personal goals added. At the moment we're getting around the public/personal issue by making goals public once a certain threshold of people have typed it in manually. This way no private goals will ever be made public (though as the site develops I'll likely have to make it a bit smarter).

Nothing big, but something I just thought I'd bring to your attention if you haven't noticed.

I'm assuming you have different colored images for the menu bar items ("Plan", "Track", etc) when users hover. It doesn't look like the images loaded immediately and it caused a couple second lag where a mouse-over caused the button to go all white.

Yes, I've noticed the lag on some of the roll-over graphics.

I'm planning on converting the images to use CSS Sprites, which I believe will solve the problem without having to resort to preloading (and, of course, make the site run a bit faster too).

Nice site. I'd personally use the word "checkpoint" or something instead of "sub-goal", but it's a small thing.

Funny you should mention that. One of my beta testers was particularly hung up on the word "subgoal". She thought it sounded too technical, but so far everyone else has thought it was okay.

There are some other wording changes I'm contemplating, so I'll have to give this one some further thought as well.

Thanks

Yeah, I got a 500 error as well when I said I really want to ... poop

I have JS enabled.

Ah, thanks for pointing that out. I figured out where the error was and fixed it. It wasn't the wording that was the problem!

But, yes, the site now accepts poop. :)

I too have Javascript. Nothing works, I get an error 500 every time.

Note: this is not browser-related:

$ curl -I http://www.goalbot.org/plan/start/sleep/

HTTP/1.1 500 INTERNAL SERVER ERROR

[...]

I have to dig further into what happened there. It should be fixed now though, I've tested through several options.

I got a 500 error by clicking one of the example links below the entry box...

Something went horribly awry with my primitive number parsing system. It should be patched up for now, I'll have to go through in detail later and make sure there aren't any holes.

I think I like it, but I'm probably not going to use it, at least now.

Understandable. I'm still working on getting enough content in the learn and plan sections to make it truly valuable without network effect from the connect section (which is a social network/forum combo).

cool idea Can we add timelines in the goals somewhere ..and i feel popups need a bit of rework..it annoys a bit initially....

A lot of details like due dates, etc. are set up in the back end but haven't been exposed through the UI yet. I'm working towards that, should be soon.

I agree about the pop-ups, several people have mentioned how annoying they are. As per the suggestion of a previous commenter I'm going to set it up so that it is controlled by the user (i.e., they can select if they've had enough instructions after the first pop up or if they want to see more instructions). I think that will help.

On a side note, those pop ups should be seen rarely by the user once they've actually set things up (though I realize first impressions are critical).

Reply on news.ycombinator.com