What to Consider Before Starting a Technology Project

I was recently advising someone how to begin a technology project. In my experience, there are four main considerations to examine before one determines an appropriate technology and implementation plan.

1) Determine practical constraints
I put this first as there are often practical factors that dictate what can and cannot be done. There's no point in wasting time with utopian planning when one has a shoestring budget.

So before beginning consider your:

• budget
• timeline and deadlines
• available human resources (skill set of staff and their availability)
• available technical resources (access to servers, database, hosting, etc.)
• level of support needed post-launch
• IT rules (there may be internal rules that require usage of certain software or against open-source or third-party code, etc.)
• other business rules (e.g. corporate style guides, suitability of data being offshore, privacy, etc.)

After going through the other steps and getting some estimates you may want to or need to revise the budget or resources needed based on more solid details. Still initial limits here are necessary to refine the scope.

2) Formalize your applicable business goals
I'm constantly surprised by the number of websites that not only fail to meet the needs of the website (such as a functioning shopping cart) let alone meet the overall needs of the business. Before embarking on a project consider and prioritize the relevant business goals. All future choices and implementations (and budgets) flow from this. If something is not meeting the business goals then why do it? It really is okay to not do something - and this can be a better option than doing something poorly. Also no technology can meet all needs, so prioritizing your goals helps determine what is most needed and what can be sacrificed or postponed.

3) Know your users and would-be users
The profiles of your customers/clients/users can dictate your path as much as practical considerations. For example, if the bulk of your customers you wish to reach don't have highspeed access (it still happens) then streaming media probably doesn't make sense.

You should know the following about your intended audience:

• platform (Windows or Mac operating system, version)
• Browser used most
• Plug-ins installed (most users will not install a special plug-in just for you, few organizations are important enough to motivate such action)
• Internet access (highspeed or dial-up)
• where they access the Internet from (work, home, laptop remotely, public library)
• mobile devices (if have one and platform)
• accessibility needs (e.g. visual or motor control issues)
• education level (this can help determine literacy level which will can guide writing style)
• technical literacy (how long have been using Internet, software, etc.)
• age (not always relevant, but can relate to other factors, such as amount of leisure time, technical literacy, accessibility needs, etc.)

I didn't include gender as I have not seen any convincing studies that indicate women and men use technology differently. So I'm not sure it is relevant - other than possibly to guide visual design or writing style.

4) Plan the individual project goals
Goals can address high-level considerations such as if the plan is to test the waters or achieve parity with a competitor, to serve one market niche or another, to dovetail with other projects or create its own buzz, and so on. With theses goals in place, it then becomes easier to determine plans for launching, roll-out, promotion, timing, support, etc. For example, it can help determine whether a phased approach is desirable or a full launch. It also makes it easier to address the daily challenges and roadblocks that arise during implementation when you can refer back to the project's goals.

Once these four main considerations have been completely mapped out, it makes determining one's path much easier. It will help narrow down viable options and make implementation plans easier. In the end, it also helps deliver technology that needs of the business and users and will guide expectations.

Renaissance ROM a Cautionary Tale

I used to love the ROM (Royal Ontario Museum) until their make-over project began. Called “Renaissance ROM”, the rebirth is like how ghastly corpses in horror movies jerk up for one last gasp before actually keeling over. As the project has pretty much wound-up now, any hope that the ROM could ultimately redeem themselves have been dashed.

So I have been seething in my dislike, not for the new architecture, whether the rather-ugly exterior and the wasted, sterile interior spaces with uninspiring, throwback to the 19th-century new exhibit presentations, but mostly for the general disrespect for museum visitors that the management of the ROM clearly has.

TVO’s Agenda recently aired a special on Toronto’s 2007 architecture (you can watch it via podcast or read the blog on their Episode page). They were inspired by this year’s Pug Awards, the awards for Toronto’s ugliest new architecture (which you can vote for online – their website is cool and effective, unlike most of the nominated buildings).

The dislike of most of the TVO panelists for the ROM inspired me to do this blog. There was one panelist who liked the ROM but even then he admitted that defending it was like “defending a child molester”.

I was thinking about what went so horribly wrong and realized that the ROM’s project problems are ones I see with website launches/relaunchs. So here are some cautionary tales for us all:

1) Don’t let the designers talk you into something too flashy
Projects of this nature do need to be visually-pleasing and grab attention, but architecture (or homepages) should fulfill its functions. A main goal of this project was to increase exhibit space, yet this project added only a small amount as the pyramids have lots of wasted space - are all sizzle and no meat. So it is design for design’s sake, but yet glass pyramids are so overdone for the past 20 years, it’s not even good design, it’s cliché and boring. I’ve seen plenty of examples of good design that serves function, but sometimes designers need to be reigned in a little (even if they are “stars”). Or if you’re going for style only, make sure it’s damn good.

2) Have adequate plans.
Libeskind submitted his plans for the building on hastily-drawn napkins – and it seems planning did not progress far from that. The Crystal is not very crystalline as Canada’s climate was not considered and the all-glass crystals were soon changed to partial glass. It’s important to do some research and really know what you’re getting into before proposing a project, let alone giving it the green-light.

3) Not thinking of your visitors first
The ROM seems to me to be too inwardly-focused, to the point that the project seems all about pleasing themselves and their peers rather than their visitors. Clearly, they decided to go with Libeskind with his cocktail napkins because of his stature (no doubt receiving external pressure to pick a star) and once this route was chosen, any desire to put users first seemed to go out the partially-glass window. The role of users seems barely considered and the design hasn’t impressed anyone I know either. The internal spaces are sterile and the pathways lack drama. Yet, it’s not just in the architecture that the ROM has forgotten its visitors, as they have turned their back on one of the major client groups, families. Families represent a significant portion of ROM’s visitors, yet they built only one family bathroom and they took away the nursing area. The kids’ area is small and dull. The dinosaur gallery is now just a “collection of bones” as one TVO panelist indicated, with the jagged, small spaces crammed with bones piled on top of each other. In all galleries, the things kids and adults enjoy (and that inspire learning) such as recreations, multimedia and interactive exhibits are almost completely gone. Signage has been kept to a minimum (because there’s no space in the crystals to hang them, I presume) and the few remaining signs are only for those with an applicable degree. I’m sure the curators enjoy this presentation - the visitors don’t. Organizations often make websites reflecting the company’s intentions and structure, but I believe a lot of this is done due to not knowing better. The ROM used to know how to be user-centric, but now instead they have disdainfully decided to create a museum for themselves and their ilk.

4) Making tasks hard to do
Here’s when you know there’s a problem – they made finding the gift store difficult. You can walk by it and not notice it at all. It’s almost impossible to find the café (not that many would want to – it’s all brie and arugula, now). So when even making-money isn’t a primary goal of the ROM, you know there’s a problem. Navigating around the museum is problematic. This could be aided by decent signage, but there’s no pathways built in to either aid finding your way or to create a sense of exploration. Building task-oriented structures, whether buildings or websites, isn’t glamorous – but when visitors can’t do what they need to, they get frustrated and the company misses opportunity to make money.

5) Not fulfilling the ultimate goal
The overarching goal for this was to make the museum a destination and thus attract more people. Instead they get mediocre architecture that isn’t awesome or unique enough to draw people (Why didn’t they talk to OCAD – now that building has drawing power) and once inside, the lack of drama or whimsy will make the building instantly forgettable. The row upon row of glass-boxes in the galleries and their family-hostile attitude will not inspire locals.

Good projects go bad for many reasons; this project is just so spectacularly bad, it’s hard to comprehend. With its huge cost and years of effort, there’s no excuse for fouling up so badly. So there’s comfort in knowing that even if you are part of a dubious project, it definitely could be worse. My ROM membership expires in two days; I won’t be renewing or likely returning.

Post-Mortem

I love the title of this blog posting as it refers not only to the standard post-project review, but more appropriately it describes my mental and physical state lately.

I really do feel like I'm back from the dead. My health is back to normal and I'm eating and drinking much healthier. I was feeling overwhelmed and under-appreciated and incredibly exhausted! I never did get my relaunch party (moments like this were made for Lesley Gore songs) but since that was cancelled I used the time to go to Toronto Island with my family and we had an incredible day.

So here comes the lessons I am passing on to Webslinger readers that may be undertaking a big web project.

Lessons learned
1) If I were to do a project of this magnitude again, I'd have the entire core team working in the same space and preferably cloistered from regular business operations - too much got in the way, distance or otherwise
2) Get a caller ID telephone - I screen and filter my emails, why can't I do that with my phone? The equipment is relatively cheap and the time they would have saved would have been invaluable.
3) Hire one extra person to do the HTML. I'm sure that that everyone who has ever done a project like this feels they need more people, but one extra person - even a temp - would have saved me some time I really could have used for more complex decisions and deliverables
4) Insist on professional testers - we recruited volunteer testers and some found some problems, but over all they missed a lot - this meant the programmers and I were testing and finding things and this wasn't the best use of time. Testing is a valuable skill and it's worth the cost to hire professionals
5) Deadlines met or else! I had difficulties getting people to hit deliverable deadlines despite knowing about things well in advance. This ended up meaning I had to work insane overtime to do the work or things got dropped. An electric cattle prod and pillory would have definitely helped.

Pats on the back
1) We hired excellent partners that offered really useful suggestions and expertise
2) I booked people's time for this project months in advance
3) We resisted the evil powers of scope attack, like Superman valiantly resisting kryptonite
4) I held focus groups which raised some good points and confirmed things I suspected
5) User testing found things that needed to be addressed but that I had not considered - it was relatively easy to do and garnered great feedback
6) I reviewed other websites - I looked at a lot of other industry and general websites and got lots of best practices and inspiration - it took some time but ended up with some great ideas, living examples to prove my points, and saved re-inventing the wheel
7) I did tons of research including reading Jakob Nielsen's book and columns, Jennifer Kyrnin's newsletter, other newsletters, attended lectures - this was also a lot of time but really meant that we knew what we were doing beforehand
8) I put our user first in devising the new website - we solicited their feedback and really listened to them. Usability was paramount as was the KISS method.
9) Great team - committed, intelligent and diligent
10) ME - I was amazing!
I programmed, designed, tested, researched, planned, wrote, edited, troubleshot, presented, trained, project managed, consulted and compromised. Yes, with much stress but also with unparalleled skill!! I have yet to meet someone who can do all these things - so the biggest pat on the back goes to me!

It's Alive - But I'm Not

The Relaunch is launched! My burden has been released. I'm free at last!

The new website is now live, but I'm anything but.

This project almost killed me. I guess I'm too damn old for the endless overtime and stress. In many ways, this relaunch is not much worst than others I've been through or heard about - just hell while you're living it.

I'm so incredibly exhausted, overly grumpy, I've had insomnia and I feel nauseous frequently. My blood pressure was at the highest ever and I'm sure I gained weight as I have barely left my chair for weeks. So this website relaunch has almost literally killed me.

I was so glad to see my daughter again I had to go many days without seeing her at all - recently I went three entire days without seeing her (during that time she was playing with her Fisher-Price Little People and the mother and child were playing in the house but the daddy was always by himself at the computer "at work").

I feel so much relief that it is done.

Today, was the big day and I knew there would be lots of issues. And there were - the two biggest, not surprisingly, were with Flash. But my coworkers and I solved a lot of things on the spot, so at the end of the day almost everything is going quite well.

Didn't get to do all the Help stuff I wanted to - that always gets left to the last and with scope attack there just wasn't time. But I still want to use DemoFuse so hopefully later this week... (the relaunch never dies!)

Relaunch Hell

Five, four, three, two, one... the countdown to D-Day is on! It's five days before the big day - the relaunch of our website!

I'd like to say I'm excited, but mostly I'm just exhausted. I'd like to say the worst is behind me, yet every day something completely unexpected and frequently horrorific affronts me.

Yes, I'm just damn sick of this website relaunch! And mostly, I just want it to be done!!!

My two-year-old daughter has been so sad that she never sees me as I'm working so much overtime (she's clinging to other dads at the park due to withdrawal). I'm exhausted as I'm literally working on this project day and night. By day and into the evening I'm at the grind, and come night, I'm dreaming about the project (which has had the upside of helping me figure out and remember things). I'm downing rum and cokes by the litres and upping my caffeine intake into hyperspace.

After weeks of beavering away at this relaunch, there's been a lot of trade-offs, sacrifices, uncovering Frankencode (a horrid patchwork of ghastly code that somehow, though barely, manages to come to life), cross browser chaos, emotional roller-coasters, and worst than scope creep - scope attack.

Yes, it is great to create something tangible and see it take shape. But man relaunches are hell!

Webslinger readers: please share your relaunch horror story so I can take solace in your agony.

And I forgot to mention, the company cancelled my launch party too! Months of relaunch hell and now no infamous launch party to look forward to. :(

My Brain is a Big Mesh

Just got back from Mesh. However, I've been so preoccupied with the website relaunch (yes I'm sorry to still be harping about this - at least the end is in site - June 18 is the go live date) that I haven't been able to sleep. So it's been hard dragging my extremely tired ass, attached to my stressed-out mind to this conference.

But I'm glad I did. Maybe it was the free Red Bull they were giving out, but despite my relaunch-withdrawal-induced insomnia, I had a great time, learned a few things, and talked to a lot of interesting people.

Mesh also had the unexpected benefit of having me home before 9pm so that I had time to blog and tag.

Spending the last couple days hearing and discussing all the cool things people are doing on the web, made me realize what a relaunch bubble I have been in lately. Not only has my dear blog languished, but I haven't tagged anything to del.icio.us in almost a month. I may love my del.icio.us bookmarks even more than you dear blog, so when I tagged Interaction Camp today (which I'm hoping my kaput ass can make it to this Saturday morning) I felt like I was reconnecting with a long-forgotten friend (a feeling not dissimilar to a friend request from someone barely remembered from highschool).

Okay, so back to Mesh.

First, I loved the setting: MaRS. Great mix of a heritage and new building.

Second, the attendees at this seemed way more friendly than any other conference or seminar I have ever been to. And they were very interesting and knowledgeable - I finally feel like I have found others in the know (I've been questing for a group of Internet professionals in Toronto and have been disappointed as some are too badly organized, some are full of amateurs and others are really only for marketing-types). So it was exciting to meet people with a similar background and passion for the Internet. Wish it wasn't just for two days though.

The keynotes, panels and workshops were good - particularly since the organizers had gathered impressive speakers both well-known and lesser-known experts.

I think the most useful things for me was the 15 Minutes of Fame that allowed innovative companies to introduce their wares. I'll be trying out at least a couple things from this and will be blogging about this later.

Mesh was a happy interlude - reaffirming my love of the Internet and love of talking about the Internet. Now back to the grind...

Website Accessibility Full of Barriers

I’m trying to make the relaunched website I’m working on more accessible to people with disabilities.

The World Wide Web Consortium (W3C) publishes standards for accessibility. The W3C have done a good job of organizing this information and I applaud this initiative. However, there are problems with their standards:

Some are no-brainers
Who’s using blinking text anymore? I haven’t seen it on any site – even cheesy ones – for 5 years or more. Although Facebook should take note of this for allowing those recent annoying flickering ads that were seizure-inducing to anyone. Does the blink tag even work any more?

Some can’t practically be done
For example, marking up quotations as such – while the {blockquote} tag is ideal for block quotes (we’re not using it for indenting anymore) and it is well supported by most browsers, but for in-line quotations the {q} tag should be used. That's fine except that Internet Explorer doesn’t support it and that represents most of our site’s traffic.

Some items are very time-consuming to do and of limited value
For example identifying the target of a link. Normally, this should be apparent from the context of the text, and if it isn’t then that’s bad writing. There are a few items that are of dubious value, and in checking out a Canadian website trumpeting website accessibility even they are not doing some of these items.

Some items require programming effort
There are some items I would like to do but require a programmer. For example,specifying tab order through forms and providing server redirects rather than meta redirects. The work that can be done simply, has the best chance of actually getting done.

It’s practically impossible to have a fully accessible site, but with some effort I think we can do some things that will help, specifically:

1. Using alt text in images. I was doing this already, though some others working on our site haven’t been. I’m going to get everyone to commit to doing a better job of alt text, particularly with graphs
2. We already used the header tags (eg. {h1}, {h2}) so we’ll continue doing that and we also use lists correctly too.
3. Not using tables for layout at all
4. Not using {blockquote} for indenting
   
5. Got rid of all that blinking text that we used to have so much of
6. Allowing users to increase the text size
   
7. Not using colours that are problematic for the colour blind
8. Recoding our tabular data into accessible tables.

Regarding the last point, I’m having a lot of trouble finding the best way to do an accessible table. Here’s what a colleague and I have come up with. Please help me out with any suggestions on how to do a table.

Accessible table???
{table class="xxx" summary="This table shows how the age of Mary and Betty makes a difference in their admission cost to movie tickets.}
{thead}
{tr}
{th} {/th}
{th}Mary {/th}
{th}Betty{/th}
{/tr}
{/thead}
{tbody}
{tr class="XX"}
{th}Age{/th}
{td}53{/td}
{td}64{/td}
{/tr}
{tr class="XX"}
{th}Admission cost{/th}
{td}$15{/td}
{td}$10{/td}
{/tr}
{/tbody}
{/table}

Please let me know if there’s anything we should be doing different.

Relaunch Frenzy

Anyone that's worked in the Internet will be able to share the pain I am currently going through.

I am relaunching the website for the company I work for. Relaunch hell is well under way!

While I am getting help, the company I work for is fairly small so as I'm leading this project, I'm also doing a lot. So the inner-most circle of relaunch hell!!! (I do have some great help, but no one ever haves enough help for a relaunch, right?)

Yet there are some real wins for this relaunch that I have been pushing for that make it all worthwhile, including:

• site architecture moved from organizational to client-focused and terminology more user-friendly
• stripping formatting code from the main web copy and making awesome use of CSS
• switching to XHTML
• adding elements of website accessibility for disabled users
• new design - with more real estate
• much more engaging, flexible homepage
• adding navigation aids, such as breadcrumbs, colour-delineated channels, awesome Quick Links feature (I blogged about this earlier)
• adding a lot of content to fill in gaps as identified by our search logs & focus groups
• printer-friendly version added for every page
• improved page cannot be found page (so users don't get that useless default 404 error)

And some other things. But I'm too tired from all this relaunch work and Facebook fix.

So forgive me blog readers if you don't hear from me until this is over in late June.

To fill in the Webslinger gap, I have a great idea...

I hear that I have blog readers but so far I think only 4 people read it. So if you read this blog, prove it!

Let me know what your tip is for surviving a relaunch!

Or let me know what last-minute things I should add or any general relaunch tips. Let me hear it - I can use the support and frankly this will be only social interaction I'll be getting until relaunch.