April 02, 2004 | View Comments (52) | Category: Our Thoughts
Summary: Trying to figure out how a project takes 15 months.
I read a story a couple years ago on the project development of Microsoft Word 1.0. Bill Gates thought that he could design and develop the product and have it shipped in 12 months. To say the least, this never happened. It took at least 3 years for Microsoft to finally get it right and have a product that they were capable of shipping. (Disclaimer: I searched for this information again and was unable to find it, so I am going on memory, which may be poor.)
The problem more than likely involved project management. Without a solid project management structure, almost any project is guaranteed to fail, even the small ones. Don't believe me? How many of you out there still have "unfinished" personal sites? We don't think about project management when it comes to our own sites, but even when we are designing something just for us, there is still a process that is followed.
In a past life, and hopefully future life, I was involved heavily in project management. First let me iterate that software project management is different than website project management (that is another entry though). This is one of the reasons why I am always curious about the thought process of a designer. I enjoy learning processes and finding ways to make them more efficient. That is what good project managers do (I think).
Yesterday, Kinja launched (why pick April Fool's Day?) and the web was abuzz with discussion on its features. I however was focused solely on one subtle aspect that was in every announcement. From Nick Denton's press release:
Kinja, a project we've been working on for more than a year, has just gone live.
Other sites were more specific and gave the project time as 15 months. 15 months is a long time for any project. Some ERP implementations do not even take 15 months. In any case, if a project is in development for 15 months there are two conclusions that can be drawn: 1) It is going to rock everyone's world or 2) The project management was a bit off. After playing with Kinja I am going to have to go with number 2.
I understand the purpose of Kinja is not for power users, but for people who would not normally be exposed to blogs. It does its job fairly well. However, I fail to see how something like this takes 15 months. I even raised this question at Kottke:
I am still trying to figure why something like this took 15 months to develop. That is ridiculous if you ask me. Were they constantly changing the focus of the project or something?...
I admit I should not have said ridiculous as I am sure there was a lot of hard work put into this project by some respectable people in the industry. In response to my statement this is what Meg (the creative director) had to say:
I am still trying to figure why something like this took 15 months to develop...
Well the first four months it was only me, doing research, interviews, requirements documentation, stuff like that. Then for 3/4 of a year it was only me and Mark, the lead tech guy, full-time. Only since last Decemeber has the team had two other full-time members.That is ridiculous if you ask me
But you don't really know the requirements, do you? You don't know the platform or the architecture, so you really have no idea how long the project should take given the resources ($ and human) and the requirements. I'm not trying to be defensive here, just want to clarify. It always cracks me up when people look at something and say, "Oh! That shouldn't have taken so long!" when they don't really know what's been built.Kinja looks very simple on the surface because of the audience for whom it's been designed. It doesn't have "filtering/ordering", it doesn't have lots of bells and whistles. But there's an amazing platform that belies the simplicity of the UI. And that's where the potential of Kinja lies. Hopefully it will be realized.
Meg is an individual who has been around the blogging scene from its inception and garners a large amount of well deserved respect so my words here may mean nothing seeing how I have been blogging for less than a year. However, I would like to point out the requirements that Kinja seems to have and find out where 15 months is coming from.
That is obviously a quick rundown and I am sure I am missing some points, but those seem to be the major requirements. The blame here rests on Nick Denton's shoulders. The guy starts a company, an internet one at that, where speed is everything. I can see how one programmer would take some time to develop all those features. I think Nick should have brought 2-3 more programmers in immediately. If Meg spent 3-4 months on requirements specs and whatnot, then surely they had to be complete enough to quickly develop something.
So in hindsight, I have to add a third reason for a project taking too long: 3) Lack of resources. That seems to be what Meg was dealing with and hopefully she understood that and let the client, Nick, understand that. That is what project managers do.
But Meg is right in saying that I don't know what has been built, of course unless I look at what is presented to me. I still believe this shows the lack of poor project management in this industry. We have too many people coming over from the software development side and trying to implement the exact same processes into web development. It does not work. This will be covered in next week.
In the end I am just going to use this entry as an excuse to introduce a new category: Web Project Management. In this category I am going to discuss project management issues from a website's standpoint. This will be helpful to everyone from the freelance designer to the designer working on a team of 20 people.
15 months...just not seeing it. Oh well, anyone here work with good project managers?
UPDATE: A lot of people seem to be taking this the wrong way. It was to discuss project management overall in the web industry, not an attack at Kinja. Kinja helped to remind me of the problems that are faced by project managers. In any case I have emailed an apology to Meg (in which she offers a great explanation in the comments of the hurdles they faced) and Nick and would also like to extend that apology to anyone else who has taken offense to this entry.
Trackback URL: http://9rules.com/cgi-bin/mt/mt-tb.cgi/195
You raise some interesting points. I've got a few comments.
#1 - 15 months is a long time. On the Web it's a very, very long time.
#2 - Web project management is very different, IMO, than software project management. With Web projects you need to be (and usually can be) a bit more flexible. Of course it depends on the project.
#3 - I project managed (as best I could) my own site redesign. Even though I'm not a project manager (although I've managed many projects) it did help quite a bit. I set a hard deadline for launch and I hit it, even though I had to rush at the end and had quite a few loose ends to tie up. I'm quite sure if I hadn't done that I'd still be tweaking. There is something to be said for just getting something up.
#4 - Timing is extremely important on the Web. I see this quite a bit with something as simple as my posts on my site. If I wait on an idea chances are someone else will "break" it. Same goes with Web sites and applications. Look at Orkut for example -- that site could very well be better than Friendster or LinkedIn, but I already (dragged kicking and screaming) created accounts there that I manage and I've got no desire to add another.
#5 - I think Kinja is pretty cool, I set up an account and like it quite a bit so far -- but they are a bit late to the game and should count themselves lucky if it takes off.
Keith (http://www.7nights.com/asterisk/)
This is why people hate bloggers. They come to grand conclusions about situations they know very little about. Some of your "conclusions" are somewhat correct, but there's not enough information available to you to support even those.
jkottke (http://www.kottke.org)
"shows the lack of poor project management in this industry."
I've never noticed a lack of that... ;-)
JC (http://thelionsweb.com/weblog)
Are you saying that this project wasn't software development? The interface may be over the web, but it is still software, right?
Personally, I think it's better to take a little longer and get the product in a very stable state than to release something that's buggy. And since I didn't hear any mention of a QA person or team, that responsibility falls on the shoulders of those non-QA people, which pushes the time frame out more.
Sure, they could have completed the project faster if they had had more resources. But maybe they weren't as concerned with the time frame, and maybe they couldn't afford more resources. There are always going to be trade-offs.
Jennifer Grucza (http://jennifergrucza.com)
I never claimed to be correct and in fact would love to hear why it took so long. Meg answered me and by her answer I got the idea that she just didn't have the resources to get it done.
Obviously you are being protective of a friend. I got not problem with that and I was not attacking her or Nick (maybe Nick a little bit). The reason people hate bloggers is because they get emotional over a topic where an interesting conversation can be pursued.
You point out that I draw conclusions that I do not even have enough information to support so I am guessing you have more information than I do. If that is the case I would love to see this information to get in the mind of the project so to speak.
If some of my conclusions are "somewhat", correct which ones? You are not helping me here and this was a post looking for answers, not more questions.
My "grand conclusions" boiled down to three different scenarios. If there were more that I missed, please let me know.
From a CEO's perspective, a project of this size that took 15 months to develop would be considered a failure. In fact, it would be lucky to even launch. Keith pointed out the time-sensitive nature of the web and it was a point that I was going to hit on in the entry, but avoided.
Denton was funding the project and I would be curious to know why he let it take so long. Did the lead programmer ever site there thinking "Damn it would be nice to have a couple more people helping me here?".
If the other aggregator type sites such as bloglines had the press power that Denton does or the ability to hire 37signals to design some of their pages, then those 15 months for Kinja would have gone down the drain. That is my conclusion.
Scrivs (http://www.9rules.com/whitespace/)
JC and Jennifer: Obviously my comment was not directed at you two since you were writing the same time I was.
In any case, Jennifer, what type of project that costs someone money is time not an issue? Maybe an OS project, but I am assuming Meg and the programmer were getting paid. The longer it takes, the more money Denton pays.
If hiring another guy (even just a contractor) cuts that development time in half, then that gives me the luxury of having my ads up faster and longer. Hopefully getting more people on board faster. And taking on the competition faster.
Scrivs (http://www.9rules.com/whitespace/)
True, you're losing business the longer you wait to launch. I don't know anything about this particular project - I just skimmed their webpage. But maybe this was considered a relatively low-priority project compared to others in development.
Jennifer Grucza (http://jennifergrucza.com)
I think you've raised valid questions, some that I was wondering myself. But to me, this is the kicker:
Why, with this much time spent, was it released as a beta? There are features (like the bookmarklet) that flat-out don't work with my browsing platform (firefox). Entries of newly-added blogs aren't mixed chronologically with other blogs. There are literally zero customization options (entries per page, font sizes, icons or no icons).
So, why not?
eric (http://gideondesign.com)
People hate bloggers?
Wow. I did not know that. Maybe I should close up shop?
Anyway, Jason, I can see your point to a certain degree. I mean why should Paul care how long it took anyway? It's not his problem.
However to jump on his back simply because he's asking a hard question (and to me this seemed more an invitation to discussion than any grand conclusions -- but maybe you didn't read the whole post?) is a bit harsh.
15 months for many projects, Web or software, is a long time. Of course there are reasons behind that, but I think what Paul is saying here (correct me if I'm wrong) is that there is a need for better project management on the Web.
Sure he opened a can of worms by using someone else's project as an example, but hey, at least he got people to take notice of something he sees as a concern.
I'd not have done it that way myself and I bet Paul is thinking he might have gone about introducing the topic of Web project management in a different way, but whatever -- sometimes you have to rock the boat. It doesn't mean your an idiot, just annoying.
(no offense Paul)
You imply that he doesn't know what he is talking about. I would guess he does know a thing or two about project management, which at least from what I gather, is the topic of this post.
Keithq (http://www.7nights.com/asterisk/)
This post goes way over the line, Scrivs. When's the last time you developed a nontrivial, public-facing web application?
It's seemed to me that for a while this site's popularity has been derived, in part, from the potshots you take at those that appear to be more popular, experienced, and/or successful than you.
I think this way because, since you *are* talking about bloggers, you could have asked Meg and Nick nicely for details, done some research, and put together an interesting writeup, a summary of which might be: "Why was Kinja in Development for 15 Months?" Instead you're just flailing.
Not a good way to get respect.
This site is becoming more of a tabloid than a roundtable or news-and-views site. Is that what you wanted?
Brian (http://joechip.net/brian/)
Actually, I was the project director, not the creative director. I managed this project. And I've have been doing "Web Project Management" for six years.
I've read your follow-up comments and take you at your word that you're asking and writing these things because you're honestly curious about the process and not just trying to be an ass. So I'll try to answer a bit.
Time to market was not really an issue. The idea was to have a small (very very small) team and "do it right" from a software development standpoint. So in terms of requirements: scalability. We didn't want to launch and suffer like Blogger, Friendster, et al. So having distributed servers, and an architecture to support that were requirements. Having such a small team meant, as Jennifer pointed out, no QA, so we were meticulous about writing tests for all our classes (and yes, because of scaling concerns, we didn't just bang Kinja out in scripting language, we used Java). And we planned on lots of other features as well, so what we've got is a platform, as I said, for the future. And that's really just a bit of it. I can't go into everything right now. I do plan on writing a lot more about the process on my site in the coming months.
Sure we could have used Perl or PHP, thrown it on one box, slapped on a UI and sucked down RSS and been done in a few months. That's not what we wanted to build and that's not what we've done.
megnut (http://www.megnut.com)
Hey, is Keithq Keith's brother? Anyways, the blog world doesn't like to have boat rock within our own world it seems, but oh well. Kinja, helped to remind me of the inconsistencies in people's methodology when it comes to project management.
I figured since I did some of that stuff that I could share my thoughts. This post was my introduction to it I suppose.
Annoying? Well, I guess if it gets the conversation going I have no complaints.
Anyways, I ask no one to believe I know what I am talking about until at least a couple more posts, but if you want to take me for my word now, that would be cool also :)
Besides, this wasn't just "some" person's project. This is a head honcho in the web world. A millionaire guy, so I am definitely not picking on Sue Blogger over there. Understandably, this style does not win me many friends, of course I think we are all too sensitive anyways sometimes.
But yeah I will be talking about project managment more because I don't see too many people (aka. none) talking about it.
Scrivs (http://www.9rules.com/whitespace/)
Brian, give me a break. Do you read the posts here?
Heaven forbid someone like the lowly Scrivs question an "A" list blogger!
Whatever. That kind of attitude pisses me of way more than the fact that Scrivs stepped over the line a bit here. Which, I do think he did.
Shame on you Scrivs!
Jas
Damn, you people post quick.
Meg, thanks for the answers. That is the kind of stuff I was looking for. I wasn't really going at the project in particular, like I said it just raise issues for me in the overall scheme of things when it comes to web project management.
Looking forward to seeing more of the underlying features that have been built into the architecture someday. Kinds of make me think of Amazon's architecture.
Brian: Come on now man. Seriously?
"It's seemed to me that for a while this site's popularity has been derived, in part, from the potshots you take at those that appear to be more popular, experienced, and/or successful than you."
To be honest, I really don't even know what to say about this statement. Potshots? I leave that stuff to Winer. Wow. Ummmm...yeah wow.
Scrivs (http://www.9rules.com/whitespace/)
I am beginning to wonder if anyone read this entry at all.
Scrivs (http://www.9rules.com/whitespace/)
Jas -- actually, my gripe was specifically that Scrivs *didn't* question Meg and Nick. :)
I mean, hell, it's a interesting topic. I'd like to know more about it. The only data point so far in this discussion that's been educational (my disgruntled, nonproductive postings inclusive, of course) is Scriv's citation of Meg's response from Kottke's site.
So it's not like this is adding any value. That's what I'm trying to say. I'm sorry if I've stepped on anybody's toes myself, but I think my point stands (for what little it's worth).
Brian (http://joechip.net/brian/)
I have to agree, emphatically, that 15 months is grossly inefficient. The only way I can think of it taking so long is if they built the site five times and totally trashed the whole thing, or if they totally reinvented a new sort of backend for it. If the backend is the impressive part, they would have mentioned it (which they have not), so it is not the impressive part.
Four months doing interviews and specs must have also resulted in a lot of things being thrown out the door, or slated for post-beta production. Did people say, "I'd just kinda like something which kept track of snippets of new posts on blogs I visit. That's it."? I'd bet that they went a bit beyond that.
With clear vision and understanding of what needed to be done, working full time with the right inscentive, this could probably be done by a dedicated programmer in a month or so. A lot of ifs, and if I had a lot of free time for the next few weeks, I'd challenge how long it took to create myself.
Aaron
Aaron, Meg offered a great response as to why it took the time it did. However, I would venture to offer some guesses that the specs were in constant fluctuation. I mean how new aggregators and similar sites have launched in the last 15 months?
With that said, I do think one of the major differences between software and the web is that with software the specs can be written in stone as with the web, clients think it is funny to continuously change them.
Scrivs (http://www.9rules.com/whitespace/)
Wow, the pride scars are getting noticeable now. People, put away your egos for a moment, and really read what is being said.
Allow me to preface my thoughts on this situation by giving you a little background experience that I have had, along with details that were given here.
I know nothing about Kinja's backend, nor will I ever pretend to do so. Let's say for the sake of argument, that the facts that I have tallied together are true.
1. 15 Months to get to BETA
That's all I know. And looking just at that, and not being able to peer under the 'hood', I will say this.
15 Months is not kinda long, in fact it could be considered failure long. Which, clearly, is why the conclusion of 'lack of resources' has been reached.
Allow me to elaborate. In the late 90's (woohoo!), I had the extreme privilege (did I just say that) of working with a small IT team on an online Stock Brokerage and Investment Banking site, from scratch. Utilizing our resources, which consisted of experts, money and technology, we were able to Alpha in 3 months. BETA rolled out about 6 months later, and GOLD was announced after only 3 months of testing post BETA launch. All in all, with vacation time, it was about 13 months of honest hard work to get our product to market. And even without peering under Kinja's hood, I can tell you that our project was of far greater complexity. However our resources were not lacking by any means.
I believe that attacking project management is something that is very hard to do, especially without all the facts. If the site was launched stating, "After 15 months of project management hell...." that'd be something. However, I think that is clear, and very apparent that resources are greatly needed to get a Web Site Project done quickly. Software is usually on a time-based scenario, but Web Sites involve content and that is usually where experts come in. At the brokerage, we had 9 Series-24 Compliance Officers/Brokers at our finger tips whenever we needed them, this proved invaluable to our efforts. It seems, from an expert prospective; Kinja had the team they needed for R&D. Perhaps man-power was lacked, which is a resource.
But jeepers, if Scrivs came out and said: "KINJA SUCKS, IT FAILED!", that'd be different. He merely said, for what Kinja is offering - 15 months seems like a long time to wait.
Colin D. Devroe (http://theubergeeks.net/)
Brian -
I've been reading Paul's blog since before he started whitespace. I'm fairly perceptive, and I've yet to note anything resembling a tabloid.
You're entitled to your opinion. I'm entitled to say I think it's a stupid one. 'nuff said.
Paul - Good to see someone outside the corporate environment discuss project management. On the other hand, it's fairly trivial outside that world since the scale is so much smaller, so you might find it less interesting than you expect. And the bit about blame resting on his shoulders wasn't worthy of you. The ridiculous bit was an opinion. The blame implies an accusation. I'd be apologizing to him, not just meg.
JC (http://thelionsweb.com/weblog)
C'mon guys, Scrivs is just trying to get some answers to questions that no one in the blogging world would probably ask. That's why a lot of people visit his site, because he "bucks the trends" that a lot of bloggers follow.
Please don't view his inquisitiveness as some sort of cheap way to garner hits or notoriety... he loves you guys!
Mike (http://phark.typepad.com)
JC: You are correct in that I shouldn't have put the blame on Nick's shoulders. I was look at it from the perspective that if a lack of resources was to blame then it is the person with money that people should point the finger at. Of course I did know all the facts and was wrong in that respect, which I freely admit. And of course an email has been sent to him.
The project management that I will be discussing will involve the corporate environment. There is only so much freelancing talk that I can do :)
Scrivs (http://www.9rules.com/whitespace/)
And even without peering under Kinja's hood, I can tell you that our project was of far greater complexity.
Colin, how can you say it was far greater complexity without looking? I mean, really? Now this isn't about ego for me, but much more good natured ribbing and astonishment.
megnut (http://www.megnut.com)
"If the backend is the impressive part, they would have mentioned it (which they have not), so it is not the impressive part."
Aaron -- ordinarily I'd agree with this, but IIRC, the site is intended for non-technical people to use... why are they going to care about all the nifty back end stuff?
15 months from concept to completion for a java-based project written to be nearly bulletproof under heavy load on distributed servers seems fairly reasonable to me, actually. Java isn't exactly a rapid development language. And keeping the dev team small makes sense, too. Less room for misunderstanding.
and if anyone else said this already... sorry. I was away from my desk for about 20 minutes in mid-comment. :-)
JC (http://thelionsweb.com/weblog)
Paul, keep posting on this topic (PMing), do not listen to the freaks who chime in but once in a while and dodge out. If anyone has been reading Whitespace for more than 2 weeks they should know that you like to open ideas for discussion. This isn't just about, "I'm right but I am opening comments to see if you can prove me wrong."
It seems to me you have built this place to be a discussion group of Web design, development, and deployment (methodology).
As an aside, I had no idea what Kinja was about until I read some of the blog entries about it. I still don't see what is special about it because apparently I have to create an account to do that. I just now spotted the "about" link. Since it is a launch, it would have been nice to see that information instead of 5 images of which one is clickable, but there's nothing to tell you that except hovering over each one.
And please, since there seem to be some pleasant people posting that have decided not to check their egos at the referring URL, my comments are my impressions. That is all. Just like yours.
Matthew Oliphant (http://usabilityworks.typepad.com)
Meg: I am not saying that Kinja's backend isn't complex. Nor am I stating that it was not done correctly. However, for what is being served, the backend would not need to be as complex as say an online Securities Trading site that manages $1M portfolios with 'Baskets of Stocks','IRAs','Single Company Stocks','Mutual Funds' 'Realtime Stock Quotes' and trades that would take place within 60 seconds.
It's the nature of the beast unfortunately, our business dealt with far greater variables.
I was in no way trying to belittle the work, from what I've seen of Kinja thusfar, Bravo!
And please don't consider my comment "Good Natured Ribbing". I am ribbing noone, I take no sides in this debate. I do take sides on the fact that people are will to scoff (spelling?) at someone's opinion of how long something should take, and immediately consider demeaning, belittlment, or a Tabloid of sorts. Crazy.
Colin D. Devroe (http://theubergeeks.net/)
Crap, forgot to add the part about the timeline of the project.
In my world, 15 months isn't bad. But I have been on porjects that went 6 months, and one that went 4 years. We deal with a lot of late-mid-range adoptors in my company. People making the transition from green screens to Web-based apps. The more we do, the less of a transition, but I've had people in usability as recently as 8 months ago that "didn't do much on the Web." They sat in front of a green screen all day running Host reports.
And given that many companies are hitting wider and wider audiences, they all are stretching their timelines due to many realities, some of which are mentioned here.
In the end, if the business partner (sponsor) is happy, the PM has nothing to worry about even if they were not on time and/or over budget. Because to run over time and budget would need the support of the sponsor. If the PM got that support it probably won't be termed a failure, as much as an opportunity to document Lessons Learned for the next project.
Matthew Oliphant (http://usabilityworks.typepad.com)
Just signed up for Kinja. Neat. Two problems I've noted... when you add multiple sites at once, the HTML that's supposed to be telling you "you added this, this, and this" shows up on the page instead of rendering. Second... boingboing overpowered the other 9 sites I added so I had to nix it to view anything else. Is it stacked by popularity instead of time or something? Anyway, it's interesting. I probably won't use it much, but it's interesting.
JC (http://thelionsweb.com/weblog)
the HTML...shows up on the page instead of rendering
Yeah, I've gotten that bug report from some folks on Safari. But I can't replicate it with my version of Safari. It's really odd.
And no, it's not stacked by popularity, it's by time. It's just that the crawl schedule isn't perfect, so BoingBoing had more new posts at once, and dumped. We're working on that.
megnut (http://www.megnut.com)
Maybe I'm dense, and sorry if this goes off-topic... but why would I want to use Kinja over NetNewsWire? It feels so much more restrictive. Working inside a browser interface in this manner just feels klunky.
I would assume that once the browsers start putting RSS trackers and aggregators into the main interface at the application level (and adding their own services for collection of popular feeds), or apps like NetNewswire start using the system level HTML rendering engines to display content, this sort of service would become dramatically less compelling. Or even useful.
Am I missing something here?
Andrei Herasimchuk (http://www.designbyfire.com)
Fun road crash going on here.
Anyone reading who has done any semblance of technical project management must be aghast at some of these revelations.
putin
"But I can't replicate it with my version of Safari. It's really odd."
I'm using IE 6 on Win2K. Actually, I'm using MyIE2, which is a wrapper for IE6... but it's not a browser issue, I don't think... probably a server thing, the < signs are being encoded < (if that works in MT comments, anyway. if not, assume it did. :-) )
And no, it's not stacked by popularity, it's by time.
OK. I understand. Seems like the purpose is less to put your own sites in than it is to surf around the categories and pick stuff you like anyway, and other than boingboing, I doubt any of the blogs I read regularly would be in there. Especially now that Paul's flaming everyone over there. ;-) (just teasing, Paul)
JC (http://thelionsweb.com/weblog)
Andrei -- from what they're saying on the about pages, it's for people who aren't technically proficient enough to use NetNewsWire. Think of it as NetNewsWire for your grandmother or something. You wouldn't be one to use it, but it would be a convenient tool for you to offer the less profoundly magnificent. ;-)
JC (http://thelionsweb.com/weblog)
Hmm. Maybe it's something that only happens when you first sign up for the account. I just logged in and added 4 more and it didn't happen. But this isn't the proper place to troubleshoot, I suppose.
And I *never* go off-topic.</XSL:BLATANT_LIE>
JC (http://thelionsweb.com/weblog)
Think of it as NetNewsWire for your grandmother or something.
Then maybe I'm truly dense... NetNewsWire seems far simpler from a user experience pov and more elegant than the browser-based Kinja, especially for non-technical members of my family.
Andrei Herasimchuk (http://www.designbyfire.com)
I didn't say it would work for her. Just that that was the target.
BTW, Paul, re: word...
8/12/1981 IBM introduces its Personal Computer, which uses Microsoft's 16-bit operating system, Microsoft® MS-DOS® version 1.0, plus Microsoft BASIC, Microsoft COBOL, Microsoft Pascal, and other Microsoft products.
9/29/1983 Microsoft introduces Word for MS-DOS 1.00.
So 2 years, tops. Unless you mean Word for Windows 1.0, which launched in 89, 4 years after the first retail version of windows was launched. :-)
reference (Google-translated Word document from MS website)
JC (http://thelionsweb.com/weblog)
Well in this discussion there is a reference to Word 1.0 taking 5 years to develop so in some way that timeline is lying. I know what I read before said longer than 2 years. Need to better my googling obviously. Of course this information was in a textbook.
Scrivs (http://www.9rules.com/whitespace/)
15 months does sound like a long time, but as Meg pointed out, "Time to market was not really an issue." I also wouldn't use the project's timeline as a sole indicator of "failure". I work in a large company where our group's job is to develop web-basd applications. I have been involved in projects that took over a year to deliver, and the common factor was scope creep. I understand that scope creep can be attributed to poor analysis, but sometimes it happens because processes have changed/evolved since the requirements were frozen. At that point it's important to step back and weed out what additional requirements are truly necessary and would contribute towards resolving the initial problem statement. This takes time if additional requirements are accepted because you have to do the proper analysis, but even before that you need to be sure everyone is on the same page, which usually takes a re-commit process. Why go through all this? Because it's better to get it right the first time and deliver a feature set that people will use, which in turn will solve their problem. Getting it wrong the first time or delivering something with "bells & whistles" that weren't needed may only contribute to the problem, which requires re-work, which potentially amounts to more time, money, and wasted effort.
I cannot say any of this was the case with Kinja, but I hope you get my point. Perhaps to the Kinja team decided it was less costly to do some of the things they did because it would be less costly to delay the release than to release it earlier and try to re-work it.
I do see your point about measuring the ROI of adding more development resources. Perhaps it could be less expensive hiring more people to finish faster than paying a few people who would take longer. However, perhaps the initial development team felt that they couldn't work as effectively in a larger team, and it would do more harm than good. Although I'm not sure about that since Java is object oriented and well suited for large development groups.
Anyway, I look forward to reading more posts about project management from your point of view and background to see how it compares to my experiences in my work environment.
Grant (http://threesixty.cc)
Not getting into the Kinja debate here (which means none of this is directed at Kinja in any way, shape or form), I'd just like to say that with regards to project management and specifically with process, most of the companies that I've dealt with in the past have had next to none. Most seemed to be flying by the seat of their pants and would rather deal with issues when they happened (aka blew up in their face) instead of being proactive about things. Hell, in the firm that I worked for I was the main advocate for getting some process introduced because I was getting tired of the EXACT same mistakes happening over and over on a weekly, if not daily, basis.
I look forward to your project management posts Paul. I think a great first topic would be how many web designers actually tell their clients ahead of time that problems will occur and how they plan to deal with them (i.e. for scope creep we use a form, etc). In otherwords, shit happens. Be prepared for it.
Nollind Whachell
Scrivs, I see where you were coming from with this one, and I look forward to more posts on Project Management. It is something that I am experimenting with, different forms of project management and project planning. Check out my blog for a post on it.
As far as the response you have to this post is concerned, I think the words you are looking for are;
doh!
Phil Baines (http://www.wubbleyew.com/blog)
What conclusion can we come to so far?
Don't ever questions meg or her boyfriend will beat you up. Hee.
thomas
Web Project Management... humph. That's a sore spot for me right now. Currently, I have the assignment to be part of a team to "tweak" a home page. Our project sponsor threw out a deadline of April 30th. For almost six weeks now, the team is still trying to finish writing a complex project charter document that will guide our work. I'm ready to scream. I'm of the philosophy that this project needed to be assigned to one person to implement a few, much needed, improvements. 'nuf said.
However, since the sponsor doesn't seem to understand Web Project Management or basic Web Development... I'm stuck at meetings listening to folks talk about something simple as if we are building complicated software.
I'm discovering that good Web project management involves helping everyone on the team to share the same philosophy about how to develop Web sites.
I love the paragraph from over at Zeldman's The Daily Report, If you're eating enough fruits and vegetables, you must not give a damn about protein, because it contains current life significance:
"And it gets harder to build an effective team if the UI designer distrusts the graphic designer, who hates the writer, who can't stand the lead developer, who looks down his nose at the systems administrator, who's convinced that the information architect is a useless hack - and don't even ask what they all think of the client."
chase (http://www.ontask.net/)
I am looking forward to project management being discussed here. This is something that has been on my mind lately cause I used to have it good, it was really bad recently and its a little more tolerable right now. The company I got my start in professional web development during the .com days was really good at it. My boss really had a good team, not cause the individuals were that strong (it did get better), but cause he managed it so well. Timelines, due dates, objectives, milestones, were all clear and the client was clear about what they were getting, when for what.
The company I work for now nearly went under due to incompetent management of the projects. All he would talk about was "systems, systems, systems, we need systems," as his company went from him to a staff of about 6. Needless to say that I am the only one left, its a decent job, good pay, and its what I love to do so I stick around. The problem I am facing is that its been years since I have been in a environment that had good project management. Throw in going to school full-time right now and I can't step up to the plate. Luckily we have a good virtual cfo that is getting things in shape, even if he doesn't understand the creative industry. The company is probably running its best right now, if one excuses some of the dead weight we haven't shed yet.
Additionally in a writing for new media class I'm in we just got the assignment to write a proposal and then redesign the school's arts tech web site. My group has an experienced project manager, but his expertise is software, not websites. Throw in a few clueless-to-web-design graphic artists, and one web monkey with 4 years professional experience, 4 more as a hobby. I've decided that I'll just focus on getting the A and not doing something I can be proud of. Most groups are 3 people, somehow we have 6, and half our team thinks thats not enough to redo a 5~6 page website in four weeks. I am used to taking PSDs from the designer and spitting out the XHTML/CSS, PHP, and MySQL for sites that size in far less time than that. Perhaps I can still learn something new, and at least help my classmates understand that there is more to making a website than Dreamweaver.
Caleb Jaffa (http://idiochron.org/tp)
Caleb: I know what you mean, man. Perhaps you should show them your skills and use some persuasion as well as some examples as to how using XHTML/CSS for the site will be much better in the long run, and will also allow creative freedom for your designers. The project manager will be pleased to read a lot of that, since coding the correct way is a lot more productive. And since you obviously have plenty of time to work on this, how about getting your group excited about this wonderful skill that you have, and aiming above the requirements? I'm sure you have thought about it a lot, but I would urge you to think about it a little more, and to try to educate your group a little bit. It's probably hard for them to realize that this is a better way to do the site if they know little or nothing about it. I'm confident, though (and you should be, too), that with a little help from scrivs' cssvault and other various pro-CSS sites, you can persuade them and make this project worthwhile.
thomas (http://gendes.elivy.com)
Wow, I'm not even a web designer anymore -- having ditched the web for grad school -- and even I feel compelled to post on this one.
First, I think the most obvious conclusion is this: though Scrivs is not intentionally being mean, picking, essentially, Kinja's launch date to start a discussion of (essentially, Kinja's) project management philosophy was not very sensitive or productive. I'm sure Meg et. al. are hoping to use these first few days to squash bugs, tweak interfaces, and generally experience that great post-project R&R/freaked-out bug-fixing insanity. It's not fair that they should also be called upon to discuss their own project management with a large audience of skeptical, disinterested third parties. No fair.
Now, that said, I've worked on a fairly large number of very late web projects. Project management is good, but in my experience what that really means is a) preventing scope creep, and b) hiring sufficient personnel. These things are more important than any high-falutin' process document. Often this means that a project manager cannot believe the existing personnel, who will maintain until the project has failed that they are sufficient. Clients are the enemies of designers and programmers, but designers and programmers are often their own worst enemies. Professional designers and programmers realize this. I think that the best-managed projects I've worked on have been led by aggressive, pessimistic project managers who reined in the speculations of both clients and coders.
Just my $0.02. And, as a further aside, I think we should all take to heart the fact that time-to-market was not a factor with Kinja. If that's the case, then I don't see what 15 months has to do with anything at all.
Josh Rothman (http://www.people.fas.harvard.edu/~jrothman)
Hey Scrivs, just wondering... have you ever done any web applications? Do you even know how much programming and testing time it takes? I suggest you actually show a proof that you can do something better rather than judge someone like that. Sure the actual programming might have taken less than a month, but getting all the ideas together takes a long time.
What I am trying to say is that there is no need to downsize someone based on how long it took them to build something.
Max
Scrivs, not sure what your past project management experience is, but 15 months isn't unusually long for a project (much less 'any project' as you exclaim), especially if you're architecting at an enterprise level with limited staff.
On the one hand, I can see where you're coming from, but well, that place really isn't so realistic. Development time is quite unpredictably correlated to project scope, environment, and team composition (definitely not linear).
I don't think you'll be able to get any wider generalizations (or deeper analysis) based on the Kinja case without knowing all the specifics, until which, any superficial comments made will be just that.
lhl (http://randomfoo.net/)
Have I done any web applications? Well yes I have actually. In Java no. In ASP.Net yes. And development time was a breeze with the proper planning. I try not to just speak out of my ass here guys and actually like to talk about stuff I do know about.
Again, I was not tearing into Kinja, but questioning. I think you two should read the whole entry and all the comments (Meg has a good one explaining everything) before you make rational judgements on me I suppose.
Ihl: 15 months is a long time, especially on the web. During the boom many companies failed because of long development times. Obviously with the amount of resources available 15 months seems more realistic, but is still a long time to me.
Meg says that time to market was not even a consideration, which made the whole Kinja discussion mute and we moved on to project management.
Scrivs (http://www.9rules.com/whitespace/)
Max: My series on project management has begun so hopefully I can prove my worth to you. I never questioned Meg's ability to manage projects. I was curious as to why it took so long. She responded politely and my questions were answered.
Scrivs (http://www.9rules.com/whitespace/)
We just brought on a web project coordinator (ok, so its just a project coordinator) but we brought her on for the reason that several of the developers were having to deal directly with the clients and sometimes resources and time were being misplaced and mismatched. Though there are still some bumps in the road with everyone being on the same schedule, it allows the developers to do what we do. She handles most of the client interaction, she handles some billing, she handles the final sign-off paperwork, but it makes everyones life much easier. The problem our team runs into is that we are still stuck in some of our old ways of doing things, and this means forgetting to copy her on important emails, taking a shortcut and calling the client ourselves, etc. Basically we all have to be on the same wavelength and if we aren't, then our workflow will be hindered.
Anyways, those are my 2 cents.
Bryan (http://www.gamecubecheats.info)
I did read the entry, and I don’t have much to say other than I’ve been part of some projects that have taken way too long for various reasons. True, sometimes it is bad project management that can take a project out of control, but other times it can be legal issues, lack of resources, or the fact that nothing like the project in question has been attempted before.
Scrivs: The MS Word 1.0 story is featured in a book called “Rapid Development” by Steve McConnell. It’s about software project management and is published by MS press. I had it as my textbook for my ISM4130 (System Implementation/Project Management) class back in college; you probably read the same book at some point in your undergrad/grad work. To this day (almost 3 years later) it’s still one of the books I refer to most.
Vinnie Garcia (http://blog.vinniegarcia.com)
Keep track of comments to all entries with the Comments Feed
©2003-06 9rules, Inc. | Thoughts Feed | Notes Feed
Amazon Stores: Powertools | Jewelry | Books | Health | Landscaping | Kitchen | Movies
#1
I guess I must add that of course if Meg didn't not ask for more people to be hired then she could be put at fault as well.
Scrivs (http://www.9rules.com/whitespace/)