- Posts tagged Web Development
- Explore Web Development on posterous
Jimdo Emerges as My New Web Tool of Choice
After spending more time than was probably good for me, I've slogged through a couple dozen tools for creating Web sites that met two specific criteria I'd developed and emerged out the other side deciding to pin my hopes on Jimdo. I thought someone else might benefit from my analysis, so I'm sharing it here. (Actually, this is just a very quick summary of my reasoning, but it hits the highlights.)
The two criteria I had were based in large part on issues I kept running into while using WordPress as my tool of choice for the past couple of years. I had decided that if I was going to find a new tool, it would have to have these two basic characteristics to make the first cut:
- Minimizing the need for me to know or use PHP and/or master relatively complex file hierarchies to do even fairly simple stuff.
- Allowing my clients to edit their sites without having to understand much, if anything, about how they were structured or how to use a sometimes bewildering Dashboard.
I started out by asking a question on LinkedIn. The thread that emerged was very helpful and I've struck up longer conversations with several of the participants outside the thread. I also made direct inquiries of a number of colleagues. I also found a great site that specializes in evaluating Web Site creation tools and spent a good bit of time there.
Filtering all that content down based on my two key criteria, I narrowed the initial list to (in no particular order):
- Basekit
- Jimdo
- concrete5
- Squarespace
- Interspire Web Publisher
- Weebly
- Yola
I actually built a sample site -- or part one a site -- in all of those tools except Interspire and Yola, the former because even though it looked pretty decent, it didn't fully meet my second criterion and its price ($395) combined with a good bit of Nettlebut about a slow and non-responsive development cycle and inadequate support didn't feel right, and the latter because it proved to be either too inflexible or non-intuitive.
Of the remaining five tools, I gradually eliminated the following for the reasons indicated:
concrete5 because: (a) although the tool is free and even open source, almost every add-on for it were priced higher than I thought was fair and would have made it a costly investment; (b) the self-hosted solution wasn't available as a one-click install on any of my hosting services even though to of them were advertised as being good hosts to use. (It turned out that in one case the problem wasn't concrete5 but rather the hosting service.)
Squarespace because the templates it offers are pretty simple and plain,which would be fine except that even minor customization requires using CSS and HTML and because simple tasks like centering images required me to set CSS properties for padding and border rather than just clicking a "center" button.
Weebly because, of all things, it does not support sidebars or any other notion of shared content across pages outside the header and footer. I was incredulous, assuming I was just missing something, but an email exchange with support confirmed it. Copy-paste maintenance of content I want on multiple pages is not going to cut it for me or my clients.
Basekit because I ended up having to make a trade-off decision between it and Jimdo. Basekit is a great tool. But it lacks a blog component, which was essential to one of my three new clients for whom I was doing this evaluation. Jimdo has a blog but doesn't support customized forms. I decided it would be easier to use something like Wufoo to create and embed a custom form into a site than to kludge inclusion of another blogging tool without losing the seamless functionality I was seeking. But I could have gone either way. I'm hoping the Jimdo guys will get to the custom form stuff at some point soon.
So I'm off to dive into Jimdo by building the first and simplest of the three new projects. I expect you'll hear from me more on this subject here over time.
Yahoo Clouds Web App Development With "Cocktails"
It's not bad enough that we developers have to make tough choices between native-platform app development and HTML5-based Web app development, now Yahoo comes along with a binge drink with which to confuse us even further.
The fading search giant has cobbled together a fairly ugly looking combination of technologies in what it calls "Cocktails". This guest column at CNET provides a glossing-over of what Cocktails is, without trying to be sufficiently specific to allow any real examination. The most salient fact we can derive from the puff piece is that Cocktails will combine HTTP, HTML5, Cascading Style Sheets, and JavaScript into some sort of development platform that they'd like us to think is really new. But wait a second. What does "HTML5" mean? While the label gets bandied about fairly frivolously, most developers who've taken a few minutes to look into it know that HTML5 is the next major development in HTML5 that will enable the design of richer application-style interfaces and interactions including multimedia.
Anyone developing Web apps using the loosely defined HTML5 technology suite would surely be using HTML, CSS and JavaScript, so what exactly is new in Cocktails?
In the CNET piece, Yahoo's Bruno Fernandez-Ruiz, the technical lead for Cocktails, says, without, apparently, any sense of irony, "Our platform combines basic ingredients that exist on the Web already, including HTTP and HTML, Cascading Style Sheets (CSS) for formatting and animation effects, and JavaScript for more sophisticated programming (both on a Web server and on individual devices)."
So, what exactly is new here again?
If you go beyond today's CNET piece, you can find out that the major differentiator claimed for Cocktails is the use of server-side JavaScript (SSJS). Hardly new. Check out this piece from SitePoint that's almost exactly three years old predicting SSJS would become as popular and widespread as PHP. That hasn't apparently happened yet but SSJS is hardly new. At best, then, it appears Cocktails is an incremental improvement on lots of platform-agnostic Web app development platforms that allows the developer to take better advantage of SSJS.
Yahoo has released two cocktails so far: Mojito, yet another environment-agnostic JavaScript Web application framework, and Yahoo! Manhattan, a hosted platform for Mojito-based applications.
Useful, yes. Exciting? World-changing? Don't think so.
Occupy Flash? Huh?
Ran across the Occupy Flash site tonight. The site is trying to mount a national or global effort to get users to uninstall Adobe Flash Player from their (primarily) desktop machines. The idea being that as long as the player has its present ubiquity, developers will continue to throw resources at it as an app platform despite the availability, as part of the rapidly dominating HTML5 standard, of open source approaches to multimedia apps.
Seems like kind of a weird idea to me. There's not a lot of HTML5 multimedia stuff out there yet; a great many sites are sticking to Flash while many are making a slow transition to the new standard. The move is premature. In a year or possibly less, such an undertaking will actually make sense without damaging the average user experience too much.
But it's a nice idea.
Reduce Fat on Mobile Web Pages
Here's a good heads-up from Website Magazine, though I wish they had extended the piece to offer some specific suggestions for best practices. As it is, we're left to figure it out for ourselves, which is fine except that most Web developers won't take the time to analyze the issue.
An Old Fashioned Site Done the Old Fashioned Way is Strangely Enjoyable
I'm doing a volunteer project for a friend that involves creating a site based around William Shakespeare and my friend's organization for promoting Shakespeare in our culture.
If you read this corner of the Web very often, you know that I'm usually a bleeding-edge state-of-the-art tools-and-frameworks kinda guy when it comes to creating stuff in technology.
But after I spent a couple of hours deciding which technology to build this site with, I decided that on balance it's a pretty straight-forward site and didn't need a lot of doodads, at least not yet.
So I decided to return to the Days of Yore when I was fortunate enough to be hobnobbing about with the folks who created the whole WWW and when a tool was an editor that was at least somewhat aware of HTML tags. I am building the site using Dreamweaver (CS4), starting from a free and fairly bare-bones template. And so far, it's been a real blast. No new architectures to learn, just HTML and jQuery JS operations, which, while a tad rusty, come rushing back like old friends to an Irish wake. There's something almost visceral about the process. Nothing between me and the site but a tool that really just tries to help rather than taking over whole parts of the project.
I'd forgotten how much fun this could be!
Pandora Jumps on HTML5 Bandwagon
According to EnGadget, one of the most popular Internet radio station/multimedia sites, Pandora, is ditching Adobe's Flash technology in favor of HTML5 as the new spec's bandwagon effect continues to pick up momentum.
Besides all the technical advantages going with the emerging standard gives Pandora, its developers also report that using HTML5 shaves seconds off page load times. On a site with millions of users opening tens of millions of streams, that represents real savings.
It just keeps getting brighter.
HTML5 Job Offerings Up 34%
According to The Inquirer, demand for HTML5 developers rose 34% in the last quarter. Citing statistics from the Freelancer site, the report indicated that there were 807 postings for developers familiar with the emerging standard in the second quarter compared to 604 in the first quarter.
Although the period gains were larger for HTML5, the number of openings is still dwarfed by the demand for 2,795 iOS and 1,702 Android developers in the same period. The only category of developer demand that showed a steep decline was Windows desktop work, which plummeted by 30%.
Game Closure Tech Helps @LostDecadeGames Port to iOS With Ease
According to a piece in today's Washington Post, the guys at Lost Decade Games have made good use of a beta release of an HTML5 game platform from Game Closure to convert its award-winning "Onslaught! Arena" game from a Chrome browser-based offering to iOS. The developers, who were experiencing extremely sluggish sales through the Chrome App Store, said using the Game Closure tools saved them 75% of the forecast development time for the conversion.
HTML5 game development is still being hampered, at least in terms of cross-platform deployment, by the relatively cumbersome and poorly understood audio support in the emerging specification. The music-inspired Lost Decade offering was disappointing its developers as they began to consider moving it to non-Chrome platforms. It runs fine on the desktop and laptops but not on portable devices where HTML5 implementations lag and are fairly inconsistent in some key ways as far as games go.
This period of time during which HTML5 implementations vary is a great opportunity for companies like Game Closure to step in and not only help developers bridge the gaps but also make a handsome sum of money in the process. Every technology gap is someone's opportunity to shine!
Apple Runs Afoul of W3C on HTML5 Tech Patents
It appears, at least, that Apple has filed for patents on two technologies that are considered part of the core specifications for HTML5 and has already been issued one of the patents.
That's not terribly unusual; companies who are members of the W3C, keeper of the specs, occasionally patent technologies that are part of an emerging standard. What is unusual in this case is that Apple has apparently not responded to W3C requests to provide royalty-free licenses for the technologies so that all who use the publicly available HTML5 stack can do so without fear of having to pay Apple royalties.
I don't know if this is a real dispute or not, but the W3C, which is normally a pretty staid organization, has published a request for prior art. Such a request asks anyone with information about technology in use prior to Apple's patent claims that would invalidate those claims to provide information to the W3C. The Consortium can then use that information in negotiations or litigation attempting to invalidate the Apple patents.
It certainly seems counter-productive for Apple, which has been part of the W3C committee working on the HTML5 spec, to try to claim a proprietary and licensable interest in these somewhat obscure technologies, both of which have to do with accessing secure information from a browser. But of course at this point we have only the W3C's position on the question. It may well be that Apple held or applied for the patents before the technologies were incorporated into the HTML5 specification, in which case the fault lies not with Apple but with the Consortium.
Still, regardless of where the fault lies, good corporate Netizenship would suggest that Apple not try to hold these technologies hostage from Web developers wishing to adopt the HTML5 emerging standard in their Web work. Hopefully, Apple will see the light before this gets too out of hand.
John Allsopp: "Hybrid Phone Apps Are BS"
SitePointe columnist Louis Simoneau had a good piece on the role of hybrid apps in the skyrocketing world of smartphone and portable device app development the other day. Featured in his article was a link to a fascinating podcast with long-time smartphone guru and conference organizer John Allsopp.
In this podcast -- and lots of other places on the Web -- John faces the issue of hybrid apps head on, opining that they are essentially dead ends that there are "a whole pile of reasons" not to engage in. His view is that Web apps built using HTML5, JavaScript, and CSS are destined to be far more successful and less prone to sudden shifts in the markets caused by disruptive technology and marketing decisions by the owners of the few App Stores out there. I've been saying this for some time, of course, but it isn't only the fact that John and I agree that causes me to recommend this podcast. Listen carefully to what he says; he offers software developers of all stripes a cogent look at the near-term future.
