Simplicity in user interface combined with the power of the what is returned equals uncommon success.
The Google interface allows complex queries with the most basic interaction.
The Twitter interface allows publication into the social conversation stream with a user interaction that looks very similar.
One interaction is asking, the other is telling.
A closed system can be a portal to openness through the network. This is a fundamental change in where the opportunities for software application development will be located in the future.
In the era of the desktop computer, an executable program needed to reside on the local computer hard drive and take advantage of the tools offered by the operating system. Access to APIs and documentation defined how open a system was. Ability to alter, or improve the system, to better support an application was a further sign of openness.
This same paradigm has been used to think about the coming age of the teleputer. Pundits and hackers cry out for access that is analogous to the desktop OS development environment. They don’t seriously attend to the possibility of a radical shift away from the hard drive to the cloud. This idea is a riff off of Steve Gillmor’s recent post.
A Short Interlude:
Upgrading software and maintaining compatibility through multiple versions on a desktop computer is one of the top usability problems of the desktop environment. The installed executable application model creates infinite complexity at the point of least understanding and ability to cope. Think about what happens when you move that complexity back into the cloud and give responsibility for managing it to the application developers. A “computer” becomes simple for the user, and as complex as the business model and developers of the application can support.
Tim O’Reilly, in his NY Times Op Ed piece, asks Verizon to open their platform in the same way that the computer is open— either on the desktop or the server. Although he coined the term “Web 2.0″ for his conference, he doesn’t seem to really understand the implications. The new path to openness is laid down by Steve Gillmor when he writes about the “hard drive” vs. “the cache.” With HTML/Ajax, Flash and Silverlight, small runtimes can be present anywhere and everywhere. The future of application development is against these small runtimes in the browser and single purpose network connected applications that make use of a subset of browser capability.
It’s an avenue to much greater user acceptance and uptake; and it removes an element of complexity from the local machine. This is how you dramatically reduce the hours of work required to maintain a computer / handheld device. Those who demand access to your computer and teleputer so they can load it up with the code they’ve written are not necessarily doing you a favor. They are probably just setting you up for a future moment when your phone will crash beyond your ability to repair it.
Resist the forces of complexity that wear the guise of “openness.” Closed systems can support both simplicity and openness via the network. Open systems support potential complexity at the device level and openness via the network. Open systems like Linux will enable closed system CloudBooks that will achieve simplicity, reliability and openness through the network.
If you happen to be passing through Terminal 3 of the San Francisco Airport any time soon, check out: Prototype to Product: 33 Projects from the Bay Area Design Community. Rushing through the Terminal to my gate, I didn’t have enough time to spend with each of the pieces. The exhibit features preliminary sketches, detailed illustrations, models, prototypes and the finished product. Every time I see this kind of approach to design I think that Web design should be done in the same way.
Designers of Web sites need to take the materials, the DOM, the semantic HTML, the CSS, the javascript, the images and links into account when they design something for a person to use for a particular purpose. Industrial designers need to know and understand their materials. Will there be a new generation of Web artists and designers who can paint in code?
A link tossed in to the stream by Joe Tennis on Twitter, stirred up thoughts about failure. Joe’s pointer was to a blog posting on the process of creating computer games, and the ideal of setting up an environment where failure can happen faster and isn’t punished. That’s a unique idea in this day and age.
It brought to mind a quote from a late Samuel Beckett novel called “Worstward Ho.”
Ever tried.
Ever failed.
No matter.
Try again.
Fail again.
Fail better.
Samuel Beckett
If you intend to participate in a creative profession, whether it’s writing fiction, making paintings or plays, creating companies, products or software— you’ll need to learn to live in, and with, failure. In a sense, success is the failure that we’ve made an accomodation with. We shoot for perfection, and we always fall short. Dave Winer summed it up in 1995 in his motto for Living VideoText — We make shitty software, with bugs. Software must ship prior to perfection, in that way it’s like life. We must live our lives prior to perfection. If we wait, we’ll miss everything.
Failure is tied to risk. If failure is not an option, risk is not an option. If risk isn’t an option, only a very small kind of success is possible. The principle is the same as an investment portfolio. You can banish risk, but you can’t expect a high level of return. Risk is a requirement of potential high return. The same is true in any creative pursuit, if you want a big success, you’ll need to learn to live with risk and failure.
And not just live with them, but to call them friends. Learning how to fail faster means learning how to succeed faster. Creating a safe environment for failure encourages risk taking and exploration. It gets you there faster. But just as with success, not all failure is equally successful. Failures need to be crafted just as carefully as successes. Just ask Samuel Beckett…
Thinking of relation between surface to depth, surface to the underlying material. These are fundamental questions when thinking about visual design. The ink on paper designer has tremendous freedom to create whatever the imagination can conjure.
The fundamental materials are the printing process and the selection of paper type. Clearly a print designer can imagine things that can’t be produced in black ink on newsprint. Generally, an experienced print graphic designer takes print production methods into account when starting a visual design. It’s called designing into the production process. When you do this, things go smoothly when it comes to to fire up the presses and put ink to paper. When you don’t there’s panic at the press check.
Of course, the fantasy is that you can make the surface manifest just as it exists in your imagination— the physical world has no claim on the execution of design. This, to some extent, is the state of much of visual design on the Web. I blame photoshop. While it’s true that just about any visual design can be built for the screen— it’s not always a good idea to do so. The horror revealed on a ‘View Source’ tells you why.
For a long time now the visual design of websites has had no relationship to the underlying structure of the document. It’s as though the visual manifestation of the website had no relation to its material. There are lots of web production processes designed this way. It’s usually done in photoshop by publication designers who know nothing of HTML or the Document Object Model.
Traditional industrial design is all about understanding your materials and the purpose of the thing you’re designing. Putting the underlying structure of the document in sync with the visual design creates a flexible, powerful whole. It allows websites to serve multiple output devices, including meeting ADA requirements.
Like the effort to get web browser makers to adhere to standards, and the new effort to get HTML email readers to abide by the same standards— this effort needs to be directed at the Content Management System makers and the visual designers of websites. CMS’s should produce POSH (Plain Old Semantic HTML), and designers should create visuals that respect the structure of the document.
Not sure what this court ruling really means, but the Register and TechCrunch are reporting that California courts are “leaning” toward requiring Web site accessibility for the visually impaired. For the folks who build websites using progressive enhancement and POSH (plain old semantic HTML), this would not be a problem. (Thanks to Joe Tennis for the link to POSH)
For the most part front end presentation code has been either been patched together by backend developers, or created by visual designers using photoshop with no understanding of how their pictures relate to code. There are a few brave souls that continue to spearhead the concept of designing with HTML. While it’s hard for the front end designer/coders to set the strategic agenda, I think it’s time they did. Right now there are a few people who can fill that role, I’m thinking of Jeremy Keith, Zeldman, Jason Fried, Eric Meyer and a few others.
I wonder if it would be a good or bad thing if the courts mandated POSH and progressive enhancement? I’m sure some back-end developer will create some inflexible, horribly tortured way to meet the requirement by creating multiple versions of a Web site. And then some online journal will document it as a “best practice.” This really could turn into a case of the blind leading the blind.
Jason Calacanis has provided us with the official definition for Web 3.0 — smart people + Web 2.0 technology, a recipe that amazingly corresponds to his own Mahalo project. People around the blog-o-sphere were up in arms, gnashing their teeth, no one had realized that it was time to define Web 3.0. Bloggers quickly polished up their definitions, counter-definitions or attacks. Some claimed to have defined Web 3.0 sooner and pointed to prior art.
But when the din resided, they asked me, although we’re not sure what Web 3.0 is, and we’re not sure why it makes sense to assign numbers to the Web— what is Web 4.0? Surely if we are going to invest our blood and treasure in the Web, we should associate ourselves with the highest possible number.
So here it is, the official definition of Web 4.0: It’s Web 2.0 mashup/api/services technology + user-asserted identity + really private, important personal information. Smart people are in there somewhere, but really— that approach is soooo Web 3.0. You may ask, can we see any of these Web 4.0 companies? Sure, there are a few starting to emerge, take a look at: Microsoft’s HealthVault, whatever Google’s Health initiative turns out to be and on the financial side, things like Mint and CakeFinancial. Although on the financial side these companies aren’t really 4.0 yet. Look for a vault that contains all your financial data which the vendors with whom you do business will be obliged to deliver to you. You’ll be putting the digital media that you own in there as well. Oh, and throw Doc Searl’s idea about Vendor Relationship Management in there as well, you’ll store your VRM prefs there as well. Stuff you are, stuff you own, data about stuff you own, stuff you want, and of course, your attention data. But it’s gotta be secure and it’s gotta solve the identity problem.
Scoble writes that Mahalo, Facebook and TechMeme will “kick Google’s ass” in 4 years. Actually, he both writes and videos his prediction.
We live in a polluted environment on the Web. We destroyed the value of the Meta Keyword tag, and with spam we’ve made email a pain and with SEO spam, we’re poluting Web search results. And our poluters use automation for the purpose of polution. No wonder the internet is boring. We’ve fouled our nest, and now we retreat to the walled gardens, gated communities and da club. The Web begins to organize itself into feudal kingdoms.
Calacanis makes products, Winer makes networked formats (and products to support those formats). Calacanis thinks about the Web’s ecosystem, but he also thinks about the economics of the Web and is willing to pay for value at the leading edge (even in the so-called social network space). That makes him rare.
So why the break? What’s really upset Winer about Mahalo? That it’s not a platform? This seems unlikely. While Winer is usually good about thinking about ordinary users, in this case I think he’s really thinking only about developers as users, and not ordinary users.
What’s good about Mahalo? Compare the search results pages for “how to speak french.”
Google: How to speak french
Mahalo: How to speak french
To my eye, the Mahalo page is more useful if a person would like to learn how to speak french. Granted the Google page lists the Mahalo page in its results, but the Google page is filled with advertising. Google doesn’t really tell me how to learn french, it provides a list of pointers based on a keyword query and page rank algorithm. Google doesn’t even know or care about my interest in speaking french.
Is the Mahalo page useful to a Web developer? Is it a low-level network protocol or format that can be mashed up into something new? Not at the moment, but it would be very useful to a developer who wanted to learn to speak french. The interesting thing about Mahalo is that it brings editorial judgement into the process of searching and finding.
And that injection of the “editor” is probably what Winer objects to the most, although he hasn’t put it that way— and maybe doesn’t realize it. Most of the technology that Dave has built is for the purpose of empowering the individual. Blogs, RSS, OPML, Manilla, Frontier — all these things give power to the individual to create. Mahalo doesn’t do that, it just provides good answers to questions. And to do that it needs smart editors to compile, structure and write answers. This makes the editor an important filter. Dave prefers to build his own filters and empower users.
From this perspective, Mahalo moves things in the other direction. Mahalo may start with good editors, but that may not always be true. If I search for “Dave Winer” on Mahalo, what will the editors return as a result? Will the break between Winer and Calacanis have any affect on the results? The human element can provide the extra intelligence that makes the difference between a good list of pointers and good answer to a question. It can also inject bias, cultural prejudice, political agendas, economic agendas, etc. Will Mahalo have the wisdom and ability to respond to its inevitable mistakes? And once that process starts, will Mahalo still be able to provide good answers to questions?
So maybe Winer is thinking about users. Calacanis has created a useful product, but he needs to answer the questions that Winer hasn’t been able to articulate.