Ask HN: Great design without designer on-board?
9 comments
Here are some tips from what I have learned as a coder that has been forced to design:
1 - find another product that you like the design for, look for something simple and with an appropriate colour scheme for your market. Use this as a starting design.
2 - Read smashing magazine for typography and layout tips.... actually just read it all the time. It's a great resource.
3 - As with code design, keep your interface simple. Initially avoid gimmicks through JS, just make something that is clean and works.
4 - Release the best interface you can to your users and ask for feedback. Some users offer great feed back others don't. Don't be afraid to ignore some feedback but most feedback has a kernel of truth to it.
5 - As with code, iterate quickly and ask for more feedback.
6 - Don't ever think you can't design, everyone starts somewhere, everyone makes mistakes. Stop procrastinating and get on with it.
7 - Sometimes there is necessary complexity. Apple are famous for simplicity but if you've ever used Final Cut Pro you will know that complex things are actually complex.
Good luck.
1 - find another product that you like the design for, look for something simple and with an appropriate colour scheme for your market. Use this as a starting design.
2 - Read smashing magazine for typography and layout tips.... actually just read it all the time. It's a great resource.
3 - As with code design, keep your interface simple. Initially avoid gimmicks through JS, just make something that is clean and works.
4 - Release the best interface you can to your users and ask for feedback. Some users offer great feed back others don't. Don't be afraid to ignore some feedback but most feedback has a kernel of truth to it.
5 - As with code, iterate quickly and ask for more feedback.
6 - Don't ever think you can't design, everyone starts somewhere, everyone makes mistakes. Stop procrastinating and get on with it.
7 - Sometimes there is necessary complexity. Apple are famous for simplicity but if you've ever used Final Cut Pro you will know that complex things are actually complex.
Good luck.
As a postscript to this I'd just like to add that I do all my 'design' in Inkscape which is free so there is no need to spend a lot of money.
I would advise any aspiring software developer to get their hands dirty with design as early as possible in their career. Hardcore programmers might deny it but interface design is as important to most software as the database structure, code architecture and algorithms. You can hardly call yourself a software developer if you are unable to develop a key part of software.
I am a programmer first and foremost (well, strictly speaking, a journalist first and foremost, but I won't go into that) and used to be terrible at design. But I worked at it, completing tutorials at sites like psdtuts.com and trying to keep up with the latest design trends at showcase sites such as webcreme.com. It was a couple of years before I could create anything I was happy with, but now I design all my software, and I have to say that it is so much more fun turning your own interface ideas into code, than doing it for someone else's ideas.
I am a programmer first and foremost (well, strictly speaking, a journalist first and foremost, but I won't go into that) and used to be terrible at design. But I worked at it, completing tutorials at sites like psdtuts.com and trying to keep up with the latest design trends at showcase sites such as webcreme.com. It was a couple of years before I could create anything I was happy with, but now I design all my software, and I have to say that it is so much more fun turning your own interface ideas into code, than doing it for someone else's ideas.
I completely agree with the comment above.
I never worked on UI all my life and when I started learning the infamous 3 - html/css/js - it opened up a whole new world.
I think I actually like it now, though it was tough initially. I know I am not stellar at good looking UI, but I can whip out "passable" designs which you wouldn't puke at.
More tips for an all dev team:
- I found the typography defaults that blueprint css has, to be sane and easy to tweak.
- Use good icon sets (psptuts, smashing magazine) - dont reinvent the wheel there.
- Read. Read. Read. - ALA, smashing mag, psdtuts.
- Practise. Practise. Practise. whatever you have read.
- Enjoy doing it. Once you add passion to the mix, life gets easy.
I never worked on UI all my life and when I started learning the infamous 3 - html/css/js - it opened up a whole new world.
I think I actually like it now, though it was tough initially. I know I am not stellar at good looking UI, but I can whip out "passable" designs which you wouldn't puke at.
More tips for an all dev team:
- I found the typography defaults that blueprint css has, to be sane and easy to tweak.
- Use good icon sets (psptuts, smashing magazine) - dont reinvent the wheel there.
- Read. Read. Read. - ALA, smashing mag, psdtuts.
- Practise. Practise. Practise. whatever you have read.
- Enjoy doing it. Once you add passion to the mix, life gets easy.
pg has frequently suggested "simple" as a substitute for "good", using Google as an example. I think this is the best you can do, in the short term, if you don't have a designer that shares your vision. Tweaking simple designs is much easier than tweaking complex ones, so if you do simple now, when you have a few thousand dollars to spend on a good designer, you won't have to undo what you've done in the meantime.
I've actually found that easier said than done, in a very large, very mature, codebase. The more functionality your software offers, the harder it becomes to make the UI simple. I've spent some time learning about design and UX in the past couple of years, diverting some of my mental resources over to the task (the only new programming language I've learned in that time is JavaScript, but I've also learned a lot about UI, user testing, making gradients and buttons, grid systems, CSS, browser bugs, etc.). There is enough programming to keep it challenging, and enough new tools involved to give me (a lifetime nerd with an affinity for how computers and software work) a relative advantage over creative folks just starting out with design for the web. I've been avoiding admitting to myself that one of my primary roles in the company is design...but it's become that by necessity. I finally broke down and bought the Adobe CS4 Web Design Premium bundle, so I have the same tools all the real designers are using.
I've actually found that easier said than done, in a very large, very mature, codebase. The more functionality your software offers, the harder it becomes to make the UI simple. I've spent some time learning about design and UX in the past couple of years, diverting some of my mental resources over to the task (the only new programming language I've learned in that time is JavaScript, but I've also learned a lot about UI, user testing, making gradients and buttons, grid systems, CSS, browser bugs, etc.). There is enough programming to keep it challenging, and enough new tools involved to give me (a lifetime nerd with an affinity for how computers and software work) a relative advantage over creative folks just starting out with design for the web. I've been avoiding admitting to myself that one of my primary roles in the company is design...but it's become that by necessity. I finally broke down and bought the Adobe CS4 Web Design Premium bundle, so I have the same tools all the real designers are using.
I could use help with design too and it's not so much that I can't design but that I'm too close to my app and need outside opinions about how to lay stuff out and how to phrase things. I wonder if startups could moonlight a little and help one another out on design, doing mockups and copyrighting with the expectation that the effort contributed will be reciprocated.
I think such an idea would be very useful, but with any such idea it is open to be abused. However, with an appropriate community behind it (this being a good example) and good implementation, I can see it working.
What concerns me is startups which wish to run in Stealth, how could someone give advice if they need to be in the "inner circle" to provide help and guidance? (Presuming a VC isn't used).
What concerns me is startups which wish to run in Stealth, how could someone give advice if they need to be in the "inner circle" to provide help and guidance? (Presuming a VC isn't used).
1) solve a problem burning enough for early adopters that they'll be willing to invest time to figure out how to use the poorly designed product
2) try to offer users just 1 action per page
3) watch logs and talk to customers to see what is/isn't being done as desired. adjust size/color to get results.
4) invest in UI pro when ready to move beyond early adopters to a mainstream audience. design does matter.
2) try to offer users just 1 action per page
3) watch logs and talk to customers to see what is/isn't being done as desired. adjust size/color to get results.
4) invest in UI pro when ready to move beyond early adopters to a mainstream audience. design does matter.
true, ui != photoshop. one can have pixel-perfect but unusable site
i like ui.jquery; adapt the API & bundled example source-code to simple projects, then after practicing, apply to larger projects.
for ornaments, i like inkscape (pkg_add inkscape or apt-get install inkscape). it's parametric, svg->png. the bundled tutorials are extremely good.
gimp and imagemagick sometimes are useful, but not necessary.
i focus on ui.jquery and inkscape (both free, open source) which don't have very steep learning curve.
i like ui.jquery; adapt the API & bundled example source-code to simple projects, then after practicing, apply to larger projects.
for ornaments, i like inkscape (pkg_add inkscape or apt-get install inkscape). it's parametric, svg->png. the bundled tutorials are extremely good.
gimp and imagemagick sometimes are useful, but not necessary.
i focus on ui.jquery and inkscape (both free, open source) which don't have very steep learning curve.
Well, the odds of having a good designer that speaks with people through colors and shapes on a startup team aren't all that great. Of course, you can always outsource design but here are several problems that I see:
1. Price. Bootstrapped startup often can't afford professional design and cheap design often is crappy.
2. Misunderstanding. A much bigger problem. As I see it, design isn't just pretty decorations on your website or nice drawings in your app. It's a direct projection of your idea through a media similar to text. If you have a tight 1-4 person team that understands the idea very well, a foreign person might fuck it up.
What do you do without a designer? The only possible solution I see is just do what you can and pray to %deity your idea and algorithms carry your through.
Is there another solution?