Posts Tagged ‘software’
Joel Spolsky, CEO of Fog Creek Software and author of the popular Joel on Software blog, wrote a post the other day that would piss me off if I was Jason Fried – Jason is one of the founders of 37Signals which publishes a bunch of very successful web-based software applications.
In Joel’s post, Where there’s muck, there’s brass, he starts off talking about how everybody has a “gnarly problem” – spending way too many words talking about bread and his childhood; which he apparently spent making bread.
Work that makes you unhappy is what I mean by “a gnarly problem.” – Joel
He goes on to say that the market pays for solutions to “gnarly problems.” Apparently, one of Fog Creek Software’s gnarly problems is getting their bug tracking product, FogBugz, to run on their customers’ own servers. FogBugz is available in hosted and “serve yourself” configurations. Fog Creek deals with the “gnarly problem” of getting FogBugz to run on their customers’ own servers because apparently the market is willing to pay for it. This is where Jason and 37Signals come in.
Earlier in the week Jason published a post titled Installable software? – a response to a question re: whether or not 37Signals had plans to produce installable versions of any of their applications. Jason’s response – unlikely. You can read his post if you want to know the details of the why. Here’s a summary of the why.
If we built installable software we’d have to spend a lot more of our time on technical support, write a lot more documentation, slow down our development process, and lose a fair bit of control over our customer experience. For some companies this wouldn’t be a big deal, but for us it would be a real drag. – Jason
I think Jason did a great job of summarizing the benefits of a centrally distributed application with cross-platform capability. I think this is the future of software. I think Jason and 37Signals made a good strategic decision to NOT offer installable versions of their applications. I make my living developing software for Windows systems and we spend a ton of time just making sure it’s going work on all the different available flavors of Windows – it’s a huge time suck. 37Signals thinks they have better things to do with their time and I agree.
Joel on the other hand, IMO, thinks 37Signals is making a mistake. That 37Signal’s customers want an installable version. That 37Signals isn’t going to grow significantly if they don’t try to solve the same “gnarly” problem Fog Creek is solving by offering installable versions of their products. Joel also seems to think they could start offering installable versions if they simply hired one extra employee – wrong.
So unless they (37Signals) deliberately want to keep the company small, which is a perfectly legitimate desire, they might eventually lose their reluctance to do things that seem gnarly. – Joel
Joel is wrong. Jason is right. 37Signals doesn’t need to produce an installable version of their product to grow. I think 37Signals can grow at a healthy pace selling subscriptions to their very functional and useful web-based software.
Joel makes a number of other comments that I’d find insulting if I was Jason. Yes, he throws in a few complements re: Jason’s design skills but doesn’t give 37Signals credit for producing software that works – there’s a major technical accomplishment here above and beyond the great design accomplishment.
The one thing that so many of today’s cute startups have in common is that all they have is a simple little Ruby-on-Rails Ajax site that has no barriers to entry and doesn’t solve any gnarly problems. So many of these companies feel insubstantial and fluffy, because, out of necessity (the whole company is three kids and an iguana), they haven’t solved anything difficult yet. Until they do, they won’t be solving problems for people. People pay for solutions to their problems. – Joel
FogBugz began it’s life as installable software. Today, FogBugz is available as a hosted solution. More people still buy the installable version over the hosted version but that’s starting to change. I think Fog Creek will see more and more of their customers moving to the hosted solution. I think Fog Creek developers will start to favor the hosted version over the installable version. The installable version will eventually go away.
I’m sure Joel – like a lot of software publishers is feeling vulnerable. Maybe that’s why he lashed out. Technical barriers to entry are coming down – it’s getting to the point where it’s pretty easy (and inexpensive) for a few kids and an iguana (Joel’s words) to reverse engineer a software application and drop it on a server somewhere. Fog Creek is better off if their customers think “installable” is a requirement – that’s harder to copy – there’s a barrier there. These days, it’s less about the software and more about marketing. That’s a hard thing for some software publishers, especially the veterans, to get their head around.
Joel should probably be taking advice from Jason as opposed to sending it in the other direction. I’m in FogBugz (the installable version) and Basecamp hours per week and FogBugz could use a little love from 37Signals. Oh and we’re still trying make time to upgrade our FogBugz installation – it’s becoming a gnarly problem for us.
Rich Internet Applications, RIAs, are the talk of Geek Town lately. New development tools from Microsoft and Adobe (Silverlight and Apollo respectively) promise to be the tools eager developers will use to build this new generation of software applications. Are geeks geeking out over an acronym again? Or is there something revolutionary coming down the road from Geek Town?
The citizens of Geek Town have been tossing the term RIA around for years but it has only recently become popular. IMO the term is still being defined. The Wikipedia page describing the term is a collection of loosely coupled facts (some simply wrong) that’s difficult to string together – I think it’s obvious the citizens of Wikipedia are struggling to define it well. My definition of a RIA is fairly general. A RIA is a software application that looks and behaves like a traditional client-side software application but it’s available from any computer with an Internet connection and the data it’s accessing is primarily (off-line mode will be a feature of many RIAs) stored in the cloud. That’s how I define a RIA now but I’m sure my definition, like Wikipedia’s, will evolve over time as well. The technology used to create the RIA isn’t important. Accessibility is the primary driver in my definition.
I think there’s good reason to be excited about RIAs and the changes that are sure to come with them. A RIA combines everything we love about the latest generation of web apps. with functionality we’ve come to expect from traditional client-side software that doesn’t live inside a web browser. Accessibility is what I love most about the web apps. I use everyday – I can access the same information in Google Calendar or Google Reader from any machine with and Internet connection. The current generation of web apps. is great but we’re coming close to reaching the limits of what we can do in the browser. I still use a number of traditional client-side software apps but over the last year the scales have tipped in favor of web apps. Traditional client-side software simply feels bulky, isolated, and behind the times from a design and look-and-feel perspective. Inexpensive hardware is also a driver behind the move to web apps. Installing tradiditioanal client-side software in the traditional way just doesn’t make sense from a convenience or cost perspective for people with access to multiple computers.
For web apps. to evolve to the next level the technology has to change – web apps. need more access – more access to the technology that has, until now, given traditional client-side software a performance advantage over web. apps. For web apps. to evolve to the next level they have to move past the boundaries, living in the web browser, creates. The days of dealing with the awkwardness of the << Back and Next >> buttons that don’t really apply in a Web 2.0 world are coming to an end. This isn’t the end for web apps. as we know them or traditional client-side software – this is the beginning of a development period that will bring the two camps together to build better software for users. RIAs will play a major role in re-shaping how we think about software over the next ten years.
Ebay’s RIA – San Dimas Demo
Earlier this week I came across a Windows Vista Ultimate package here in the office and discovered something interesting – a U.I. flaw in what was probably an uber-expensive package designed by uber-packaging experts from all corners of the world.
I saw the package sitting on a co-worker’s desk and was immediately interested. I’ve worked on designing software packages in the past so a package like the Windows Vista package that is obviously going to be well funded and the product of hundreds of hours of experience deserves some attention. I picked up the package, looked at the front, looked at the back, and proceeded to open it. To my surprise, and some embarrassment because my co-workers were watching, I had a hard time getting it opened. I don’t remember how long it took me to get it opened but I do know it was way longer than it should have been. Even after I got it opened, using what appears to be a last-minute hack, I had difficulty opening and closing the package. Is this for real? How many of these packages is Microsoft shipping? Did they test this at all?
The hack I referred to is illustrated in the photo I took below. The little red piece of tape attached to the top of the package can’t be part of the original design but without it I would have been really stuck. I’m sure the last minute hack was subject to protest on the designers part – it’s ugly. Did Microsoft "Patch" the Windows Vista package? We’ll probably never know.
By the way, I passed the package around the office for a little in-house usability testing and a lot of really smart people turned red trying to get the Windows Vista package opened in front of a crowd. That’s bad design – period.
Has anybody else had this experience? If so, comment.
Does anybody know who designed the package? If so, comment.
I found this post on the Windows Vista Team Blog that, after having seen the package for myself, is almost comical.
Source – Announcing New Packaging for …
I’ll admit it; I’m a Windows guy with MAC envy. The last Apple I owned (actually my parents owned it) was an Apple IIe. I learned BASIC on that Apple, battled fiercely in Droll, and huffed-it through the mines of Lode Runner; but I haven’t had an Apple anything in my possession since.
Why? I think I’m just too practical. I however, unlike a lot of Windows users, am not afraid to admit that I have the occasional bout with MAC envy. For example, I was terribly jealous about not getting to take part in the release of Panther this week. Bloggers everywhere are talking about it and it looks cool. Apple knows how to put a nice looking U.I. together and that’s at the root of my MAC envy condition.
For those of you, that like me, experience MAC envy from time to time; try sprucing up your Windows U.I. with a little product I found called MobyDock. MobyDock is a freeware application that puts a launchbar similar to the one introduced with MAC OSX on your Windows Desktop.
I haven’t actually tried MobyDock for myself yet but you can read a full review on the Lockerknome site. I’ll post a review after I’ve had a chance to use it. So, for now, deal with your MAC envy by putting MobyDock on your Window desktop. Or, make that Switch Apple is always talking about.