114. Forum

Paul Boag

On show 114: Should designers stick to designing? What goes into a usability test script, and we talk to Alex Mogilevsky from Microsoft about Internet Explorer 8.

Download this show.

Launch our podcast player

News and events

Before we look at the news I want to share a bit about boagworld. There is so much happening with both Boagworld and Headscape. If you are interested read my random news post.

However, for now I just want to mention the forum. For a while now the forum has been experiencing technical difficulties. Fortunately that is behind us and I want to encourage you all to check it out.

I can honestly say that the boagworld forum is the best forum I have ever been involved in. Not because of anything I have done, but because of the awesome people on there. They are so amazingly helpful and I have never seen a flame war or newbie bashing.

Please take a few minutes to sign up. Even if you don’t regularly participate, you can at least add your location to our listeners map :)

Instant mobile sites

I want to mention a website called MoFuse.

I have talked before about the importance of the mobile web. With a new generation of internet enabled phones we need to “moblise” our sites quickly.

MoFuse makes the process painless. It takes your RSS feed and tailors it for different mobile devices. For example an iPhone user would receive a different experience to a WAP user.

As well as RSS you can add static pages and customise the design. It is an impressive service and the basic setup is free.

I should stress that this is not an ideal solution. It does not replace a carefully crafted mobile site. However, it is a good starting point, especially if all you want is a mobile version of your personal blog.

And yes, I did create a mobile version of boagworld (m.boagworld.com) that allows you to stream the podcast on your iPhone.

Layout, do’s and don’ts

My next news item is a post by Andy Rutledge on bad layout conventions. He highlights two problems.

The first is three column layouts where the main content is framed by two side columns. Andy argues this creates several problems and demonstrates how things could be improved by taking the Apple store (which uses this layout) and redesigning it so that the two side columns are to the right of the main content.

The second problem is organising different sections of information into long vertical columns of varying lengths. This requires a lot of scanning and scrolling. Again, Andy demonstrates the problem by fixing it on an example site. He uses a horizontal approach where the sections span the width of the page rather than being sorted vertically.

At the same time Andy posted this article, the guys at Web Designers Wall posted some inspirational grid based designs. Interestingly I saw the Web Designers Wall post first and was inspired by the list of sites. However, after reading Andy’s post I returned and noticed the problems Andy has identified. Suddenly my opinion was changed.

I recommend checking out both posts. They will change the way you view layout.

jQuery for designers

At the beginning of the year I said designers should learn Javascript. I have also said in the past that we should avoid frameworks and learn Javascript from scratch. However, Christian Heilmann challenged this assumption in our recent interview.

This week I found a tutorial aimed at introducing jQuery to designers. What I read blew me away. Not only is jQuery small (approximately 29kb) it also reduces the amount of code I need to write considerably. It will quickly earn back those extra bytes.

However, what I love most is that it reads like CSS. To find elements by tag, class or id you don’t use getElementBy as you would in Javascript. Instead you specify it in the same way as CSS using # . and tag names.

Finally, it makes animation easy. You can slide things around, change their opacity and much more. Check out the extensive documentation to learn more.

Although I haven’t used it on a project yet, it is looking hopefully.

Back to top

Feature: User testing scripts

I have talked before about the benefits of user testing. I have discussed how to user test on a budget. Now, I want to look at some basics that go into every usability test script.

Back to top

Expert interview: Alex Mogilevsky

Before we get to the interview with Alex it is worth saying that he contacted me a few days after this interview saying that it could be out of date by the time of broadcast. He was of course correct as last night Microsoft announced that IE8 will be standards compliant by default. Although I still do not have a problem with Microsoft’s previous position, I know a number of people disagree with me. It is encouraging to see Microsoft listen to the community and change its approach accordingly. I am sure they will have to endure criticism from other quarters for this decision and we need to encourage them.

Paul:Okay, joining me today is Alex from Microsoft. Now Alex have you noticed how I avoided trying to say your surname there? How do you pronounce it? Ah well I haven’t got a hope have I? I’m doomed. But it’s good to have you on the show anyway. Now it’s really nice of you to come on. We thought it’d be good to get you on the show to talk a little about IE because I understand that that’s what you’re currently involved with. But before we get on to that can you give the listeners a bit of a background as to who you are and what you do and kind of how you ended up being involved with IE

Alex: Uh yeah I’m at Microsoft for about 15 years now and I worked on all kinds of projects from word to publisher and IE now. I first worked on it at the end of A5 so it is my, technically it is my fifth version of IE although the real big one that I worked on was IE 5.5.

Paul:Oh okay.

Alex: I have a reason to not like .5 version (of IE) because that was really a big project for me and I have to explain that it was 5.5.

Paul:(Pauls laughs.) Yeah it may be a .5 version number but it was massive to you. I get the idea. Well obviously um part of the reason that we got you on the show today is to talk about version targeting. But we also want to talk about a lot of different things as well. How you see the browser developing in the future? What the futures of browesers generally is? But let’s kick off with the subject of version targeting. Did you expect the kind of level of reaction that you got off of the back of this?

Alex:It was surpising to see huge negative reaction from people being published. I don’t know there were probably hundreds and hundreds of comments on IE blog. It’s hard to find the positive one. When you talk to somebody who is actually creating content for the web I get a little time to explain the concept is, everybody agrees that well there is some content there, there has to be a way to figure out what the author was thinking when they were creating it. So you should be kind of marking that content with some clue for the browser to render it properly. It is quite amazing to me that people who disagree with version point fingers at Microsoft. That it is totally Microsoft’s thing.

Paul:I mean it’s interesting isn’t it, cause one of the reasons that people said that Microsoft introduced versioning was because there was such a kind of kick back over IE 7 breaking the web, as it has been put. Um Would you think that IE 7 has been perceived as a failure within Microsoft? Many people seem to have thought of it, you know, there is this perception that is the way it is perceived in Microsoft. Is that actually the case?

Alex: Well. It depends on how you define failure. I think from one point of view it didn’t do enough from another it did too much. So go with what everybody ask us. We go and do more standard, and we didn’t get standard enough. It’s not 100% of standard but the little that we did broke so much that we really learned the hard way that we can’t make any progress without doing something about not breaking exsisting web.

Paul:Do you think that with the kind of issue, you know, one of the main comment that I’ve read about the version targeting is the fact that by default um, that you have to kind of opt into version targeting in effect to opt out of it. That by default it will render as IE 7 for time and memorial. Do you think that that’s likely to change before the release of IE 8 or is Microsoft fairly set on that now?

Alex: I don’t think we have any other option. What would that option be, is go and visit every website on the web and opt it out and then carefully figure out which ones are actually designed to work with IE 8 even though they couldn’t have possibly known what it means.

Paul: So do you think that it would be too much of a burden on people out there creating websites, that when they’re seeing a problem and they’re saying why does IE 8, you know, break my website. Telling them to just add another single line of code, you know, do you think that’s asking too much?

Alex: Well I think any assumption like assumes that every site out there actually has a person who understands why it is what it is and is actually capable of fixing them. The vast majority of people publish and they don’t know what a version is. They don’t know if their site is in Quirks mode or standards mode. They look at this as a browser bug. But, maybe there is something that is relying on a bug in previous browsers where old browsers were compatable in being non-standards.

Paul:I’m in agreement. I think you’ve come up, I mean it is a compromise what you guys have come up with, but as compromises go I think it is a very good one. I think it’s much more important to keep the web an environment that anybody can put information up on without having to be an expert in web design. If that means as professional web designers we’re slightly inconvenienced that seems to be a price worth paying from my point of view. But let’s move on from version targeting cause we’ve talked about it far too much on the show. Let’s talk a little bit about the process of developing a web browser. I think for a lot of web designers we’re entirely on the side of, obviously, you know putting things into a web browser. Getting things to render correctly. How do you go about creating a web browser? What’s involved? What should we be aware of?

Alex:Well a web browser is an enormously complicated thing. There is a reason that new browsers don’t show up too much. You have to build at the same time a decent editor, a rendering system that can render almost any kind of text and graphic. It has to be fast. It has to have object model. It has to parse a variety of different things. And is has to be compliant with a standard which is a book of at least 2 or 3 inches thick if you count HTML, CSS, and object model.

Paul:You say that developing a browsers is hugely complex and obviously it is, and that’s part of the reason that browsers don’t come around very often, or updates to browsers. But, on the other hand we do seem to see other browser manufacturers moving much quicker than Microsoft. Why is that? Why does there seem to be slower movement out of Microsoft than elsewhere.

Alex:I’m not sure that much quicker is the right way to describe it. The reason a lot of browsers are way ahead of Microsoft in a particular standard support is I think for two reasons. First we actually care about standards and we are nice people.

Paul:(laughing) You’re implying that no one else is nice and no one else cares about standards. I’m sure that’s not true.

Alex:No not really. This is how this happened. When we started working on browsers, which was, IE 1, 2, 3, 4 in the 90’s, there wasn’t so much of a standard. There was some idea you can mark text with tags in angle brackets and it will render somehow. So a b means bold. Eventually some kind of defacto standards developed. Every site was built for Netscape and it was important to be like Netscape. It was pretty much a standard. So when we built a browser at that time we were trying to be as compatable with that standard as possible. Then what happened, Netscape just went away for a few years and we were left alone. There was nobody to compete with. That is a hard place to be. You have to be compatable with the web. You can’t break anything. How do you make progress at that time. There was no versioning. At that point we decided to be nice and return the favour to other browsers and we went away for 5 years and were sitting for 5 years on the beach and letting everybody else catch up.

Paul:That was really generous of you. That was a wonderful thing. That’s a good answer I like that Alex.

Alex:Now everybody’s caught up we can be in the game together.

Paul:Okay. So now that everybody’s at the same level you’re going to go for it majorly. So that brings us quite nicely on to the future of browsers. You know, particularly IE 8. Obviously you can’t really tell us anything that isn’t already publicly known and I wouldn’t expect you to. But, not everybody that listens to this podcast already knows all that is publicly out there about IE 8. So tell us what kind of direction you’re going in. How are you going to compete with these other browsers out there now that they’ve caught you up?

Alex:Well the biggest thing to know about IE 8 is it is back. We have a real team of people who is really serious about building a good browser. The biggest investment in this one is obviously standards. That’s the biggest thing that everybody was unhappy about with the previous versions. We get as much as we can of good standards support right here. We are making major investments in CSS 2.1 doing whatever it means to be compatable. Interestingly enough it has to be defined what it means to be compatable, to say that you are compatable. We are doing as much as we can in this area. There will be a lot of exciting user interface features there. But I don’t think I can tell about that because I am really a core technology person and I learn about our UI implements from demos.

Paul:Fair enough. So, internally are you working towards a deadline of any description over this? I presume we’re not talking about another 5 years here? We’re talking about a much shorter period of time.

Alex:Definitely it is way closer than 5 years.

Paul:(laughs) That’s good. You want to narrow it down any from that or am I pushing my luck?

Alex:Yeah I don’t think I can have a number. Unless it is publicly known and I haven’t checked in the last couple of days. But you are going to see it pretty soon. I am going to set your expectations pretty low. Don’t expect this to be done in quaters because what we are doing with new standards layout it isn’t crazy and insane trying to rewrite things like this in less than 10 years.

Paul: So you’re suggesting that before too long we might see a beta. You’ve obviously come under a bit of an attack from Opera and they’re have been several comments about that. And then you’ve come out and said that you’ve done that acid test. How big a deal really is that in the real world? Do you think it is a fair test of a browser?

Alex:Well it is an acid test. As I understand in chemistry it checks the presence of a chemical, right? I guess it is a variant like that. I spent days taking this apart and figuring out what every line means and what features do you have to have, and how are they supposed to work to render a particular pixel in a particular place. It is quite and interesting test in the way that it tests very few features but it test the features in corner cases of intersting features of a standard. So you would assume if you get that far, if you manage to render text at all and you render Acid 2 you are in really good shape.

Paul:So is is fair to say…The danger is, and I have heard some people say this, with IE 8 you’ve just been going after being able to say yeah we pass the Acid 2 test. You could kind of develop those things in isolation. My hope, and I suspect your hope as well, is that you’re going to go beyond that. It’s almost inconsequential that you passed the Acid 2 test. It’s part of a broader aim to introduce better standards support. Is that a fair comment?

Alex:Yes it is. It probably would be not unreasonable to have another browser that doesn’t have an particularly promise but passes the Acid 2 test. We are really trying to be as close to standard as possible.

Paul:Let’s broaden it out cause we’ve kind of picked on IE quite a lot here. Let’s broaden it out to talk about where browsers are going generally in the next few years. I mean, what do you think we’re going to see happening in browsers generally?

Alex:Well from my point of view as standards and layout, I see actually multiple browsers being 100% compatable with CSS 2.1 which most helps both browsers and standards to really get into the new steps. We’ve heard a lot of criticism on standards not moving. Like CSS 3 some people are saying that trying to make sense of all the different parts of CSS 3 what is the next step there. In my opinion I am as unhappy with CSS 3 not moving as designers are. For this to move you have to have a solid ground of the previous version not only being a solid standard but also being supported properly. Then addding every little feature to CSS 3 become important because now we can have several implementations of that. As far as standards go I am really looking forward to the world after IE 8. And there are all the different directions of being mobile. Bringing more of the technologies into the browsers that they can perhaps take a bite out of flash or out of .pdf. Where you can have real content that was unthinkable for a browser a few years ago. Into every place that has a good browser. So I’m really optimistic. One thing to understand about developing a new browser is it’s exciting who writes is how has it. But for a web designer it is like making wine. Something happens today that will actually make the difference in the life of a web designer 10 or 15 years from now. Whatever is being shipped today, everybody you care about now has this one.

Paul:What we need you guys to do is drop support for IE 6 really quick cause then we can use it as an excuse to drop support as well.

Alex:This goes back to versioning a little bit. There is 3 kinds of versioning. There is versioning of a standard, version of a browser and version of a document. The version of a document is something that we are not going to drop ever. If something is written for IE 6 today 100 years from now we will still want to render it this way right? We wouldn’t go to the library and throw away all the books that are older than 100 years.

Paul:That’s a very good point.

Alex: In this sense many pages today are in Quirks mode. Google is in Quirks mode. If you try to render it in 100% standard it will look a little different. It will look quite odd by the way.

Paul:That’s a whole other discussion. Google and it’s standards support. What about typography? Are we going to see some improvements on that front? I mean we’ve noticed some of the other browsers starting to support web fonts and things like that. Can we expect a similar thing from IE?

Alex:IE supports font embedding for several versions now. There was an extremly complicated discussion within the standards group on how do you create a standard for downloadable fonts that can be standardized but is also friendly to font manufacturers. To people who own fonts. It’s easy to implement a hyperlink to a font that will be downloaded and used. Then it encourages to just put fonts out there, which you’re not supposed to do.

Paul:Obviously this isn’t down to Microsoft cause this is the whole standards group discussing this. That strike me as no different than what we do with imagery at the moment. Images can be pirated and that’s somebody’s intellectual property. It’s kind of just the way things have been. Do you feel that their is a kind of tightening up on things like that. Like people are more concerned about that then perhaps they were in the early days?

Alex:It is a good comparison of the difference of, whats the difference of an image or a song or a movie or a font. It is just so much more expensive to build a font. It is so easy to share if you just put out a file that if we make it easier to share we will just business model for a lot of small design companies who just build fonts. So we have to be very careful on how exactly it is being done. I think we are fairly close to be able to come up with a workable standard. We’re just not there yet.

Paul:Okay. So that is something that will probably turn up at some point but not necessarily in IE 8.

Alex:Well if that standard is embedded OpenType it is already there.

Paul:Yeah that’s true. Let’s just touch on briefly another kind of criticism that really hasn’t been aimed at the IE group but Microsoft as a whole, which is this whole issue of Office. The latest version of Office. I know a lot of people who are passionate about HTML emails who were deeply disturbed to see that the latest versio of Outlook doesn’t use IE’s rendering engine. Can you give us any kind of insight into what happened there? I mean obviously it’s not your area but it does affect you guys to some degree.

Alex:I wasn’t part of this decision but I can very well understand how that happened. When Office 2007 was being planned IE team pretty much didn’t exsist. We completely rebuilt the team for IE 7. I am not joking that we were 5 years on the beach. IE team stopped exsisting after IE 6. People went off doing their own different things. Of people working on IE 7 and now IE 8 there is litterally a handful of people who worked on IE 5. Maybe even less. It is a brand new team. When Office 2007 was being planned in 2003 there wasn’t anybody to talk to. The biggest concern about HTML email is security. So even is somebody wanted to build in IE 8 currently in a product to read email it would put a major requirement on IE to be able to be able to open all of these emails securly. If you take browser security and application security for things like Outlook it is a multiplier. It is way harder to build it that way. So I think that is totally sucks that you can’t send HTML email but I think that they decision that they had to make in Office 2007 was really difficult.

Paul:Interesting stuff. It’s been great. I’ve really enjoyed chatting with you Alex. I feel like I’ve got a little bit more of an understanding of what’s gone on inside of Microsoft and where things have gone. It sounds like a big part of all of this was your own success. That because other browser manufacturers disappeared and you were kind of left alone there wasn’t the kind of investment internally within Microsoft to keep things moving forward and it’s had ramifications. It’s very interesting.

Alex:Yeah. Thanks for talking to me. I have to say that you have a great podcast. I’ve been learning a lot from you. You don’t even imagine how much you yourself have on the future of internet. Just putting together this show.

Paul:(I can just see Paul’s head growing) Oh well thank very much Alex. I’m sure I’m not that important really but I’ll pretend I am. Thanks very much for coming on the show. We’ll get you back in the future. Perhaps once IE 8 is out in beta. You can talk a little bit more then. That would be fun.

Alex:I would love to. Thank you.

Back to top

Listeners email:

Implementing RSS

Our first question is an audio question about RSS. It asks what tools exist for adding RSS to a clients website.

The simplest solution is to use a blog. Whether you use Movable Type, Word Press or Blogger they will all generate an RSS feed. Blogs are also easy to update and can be customised to integrate with the rest of your site.

However, a blog might not meet your clients requirements. For example, they might wish the RSS feed to link to a non-blog page or even a third party site. One option is to use a social bookmarking tool like Delicious. This creates an RSS feed of all links added. Alternatively you could use a desktop application such as FeedForAll. FeedForAll is a cross platform application that allows users to easily create, edit and publish RSS feeds. It also comes with a PHP script that converts RSS into HTML via PHP. This is ideal for embedding RSS into your website.

Hope that helps.

Should designers just design

Our next question comes from Travis…

I have recently been asked to join a web development/design team as their full-time UI designer. My responsibilities would include the UI design only and not css, html ajax, etc. I know the basics of css and html and have designed standards-based sites from scratch, and am eager to learn more code to further my skill set. I have been discouraged from learning code while in this new position, and focus on the UI design. Do most web design firms share this same structure?

Excellent question Travis and one that causes much lively discussion. Infact Cameron Moll asked the same question and the last I saw there was 114 comments!

My answer is that designers should be able to code. Personally, I don’t believe you can be an effective interface designers without an understanding of the underlying code and the constraints of HTML and CSS. I respect those who say that design shouldn’t be constrained by technology. However, in the real world its more complicated.

Setting that debate aside I would suggest that not knowing code is going to damage your career. Personally I would not hire somebody who cannot build as well as design a site. Consider your next job and whether you can afford to stay somewhere that actively discourages you from expanding your knowledge of the web.

To leave an audio comment for the show skype “boagworldshow” or call +44 20 8133 5122.