Designing for all users, by Jeffrey Veen

Designing for all users

In a post about the new Google Code Project Hosting, Marshal Kirkpatrick of TechCrunch writes:

"So far the whole thing looks far less user-friendly than SourceForge but if it's targeting a technical audience exclusively that might be of less consequence."

I'm not very familiar with Google's new entry into this space, though I've met the guys working on it. And I'm glad TechCrunch mentions user experience when evaluating it. That comment, however, fails to acknowledge that product teams need to consider all users to be successful.

I've always been frustrated with SourceForge exactly because of this issue. The problem for me is not the repository; usability of managing distributed projects doesn't really affect me. Rather, it's that so many interesting open source projects use SourceForge to distribute their software. And that task - simply finding the latest release to download - is really confusing, especially for people with no experience with code repositories and the shorthand used for these projects.

It's a delicate balance, of course. Developers need quick access to the technical guts of a project. But for projects that are ultimately intended for a broader consumer audience, these sorts of sites generally don't suffice - which is why so many groups create entirely separate distribution sites.

I can't help but think, though, that there may be a middle ground that works for everyone.

This entry was written by Jeffrey Veen and posted 31 July 2006 at 9:27 AM. It was filed under Web Design. | View blog reactions

Comments

1. On 31 July 2006 at 9:52 AM Jan Brasna wrote:

Absolutely. The first "geek" piece of SW that got it right from my point of view then was Firefox with its redesigned task-focused web site and quick, easy and understandable access to the right distribution package the user needed. It's one of the most important features/functions of similar sites, it should definitely get more focus.

2. On 31 July 2006 at 9:59 AM Matt Haughey wrote:

How could anything on earth be less user friendly than sourceforge?

I've been meaning for months to make a screencast mocking how unusable sourceforge is when you want to download something. I noticed recently they actually started color-coding links that said "download". But you still have to go through 3-4 screens to get the code you want, so it doesn't help much.

3. On 31 July 2006 at 10:23 AM Matt wrote:

When I moved WordPress downloads off SourceForge they shot up 2-3x within days.

4. On 31 July 2006 at 12:35 PM Per Djurner wrote:

Absolutely agree, I never even considered SourceForge as an option for my open source project, it was just too overwhelming.

I signed up for Google Code Project Hosting the minute I saw it though and really like it so far.

5. On 1 August 2006 at 3:58 AM Michael Heilemann wrote:

Amen brother! I've always wondered why anyone would want to use Sourceforge as the base of their project. I'm pretty proficient with the internets and I'm still confused when I need to download something off of Sourceforge.

We moved K2 onto Google Code Hosting this weekend, and I'm already glad we did. Someone calling Google Code Hosting less user friendly than Sourceforge really puts their proficiency into question.

6. On 1 August 2006 at 6:17 AM Lee wrote:

Yeah, it's the classic "design for experts" approach. Sourceforge makes it really hard to just work out how to download the software (although looking at it again it seems to have improved in that regard - it used to take a good five minutes of confusion). So many projects use Sourceforge that most of the people looking for downloads aren't experts.

The lack of visual structure in the page layouts can be pretty confusing too in my opinion.

7. On 1 August 2006 at 6:35 PM Binusha Perera wrote:

hi all

One of things I guess when we are working on developing MenuTree project (http://www.menutree.net) was we followed a similar path to 37Signals. We set a source code control with subversion so that we could manage the source code. Even though it was technically challenging we manage to get things sorted pretty quickly.

A brief overview of the project ...

The objective of MenuTree is to provide a facility for people find take-away (out) fast. We were in the situation we were getting a load of junk mail coming into our mailboxes and when it comes to finding a good take-away when you need one we could never locate that handout. So the idea was born that many people may be in the same boat as us so we decide to work on it expand it and help others in the quest. Ultimately we want to reduce the paper wastage and provide an alternative way for restaurants to manage this, however said that most of our challenges are not in the technical aspect but the business side of things. A classic scenario we need to over come is the time factor that restaurants may have in doing this kind of work. Hence we decided we would rely on the community out there to help us out with it.

However even though the project is not developed with web 2.0 (though using Google maps and word suggest) myself and Ross are looking at getting the basics right. Studying our customer and ensuring that the site is simple to use and meets standards (working on this), unfortunately like some of the larger companies or the smaller midsize companies we don't have the luxury to spend time (we are working during day time, if anyone wants to sponsor us please feel free to get in touch with us :o) however we are passionate about our solution and will continue to put effort into developing it.

We are relying on the community to help us out to develop this into a better solution and provide a useful service. I am working toward making the solution more user friendly and making simple as possible for people to look for menus. We are starting small you could say tiny baby steps as we are mainly working with Canberra (Australian Capital City).

So if you get a chance we would love to hear from you and get some feedback as to how we can better provide great service.

Thank you muchly :o)

Binusha

8. On 3 August 2006 at 1:54 AM Lea wrote:

Heck, I *am* technical and *I* sometimes go round and round trying to find files on sourceforge!
(and I thought it was just me...)

9. On 20 August 2006 at 12:06 PM Dan wrote:

I just noticed (and was angered by) this myself... I downloaded something today from SourceForge, and found that the link that said "DOWNLOAD" went to a page with more information, that included another link that said "DOWNLOAD", which went to a page with a list of mirrors... finally, after the third click from the original page, the download started (by Javascript; there would have been a fourth click required if I had disabled this). Shouldn't "DOWNLOAD" *mean* download?