A couple of weeks back, Sun’s Tim Bray wrote an interesting little piece reacting to (yet) another lambasting of Wikipedia by an otherwise intelligent Andrew Orlowski. The part that interested me, however, wasn’t his defense of Wikipedia: I’m frankly more or less totally unconcerned what would-be critics of Wikipedia think of it, or the doom they perceive within its success. Why? For me, it just works, but more importantly, a substantial majority of my non-technical friends and family are avid users of Wikipedia. I can’t imagine one of them being persuaded by Orlowski’s arguments, and I’m not therefore particularly concerned about the future of the online encyclopedia; demand, in my opinion, is the surest safeguard to its future.
No, what I found most compelling within Bray’s defense was his tidbit on length. Here’s the quote:
Anyhow, in among all this tilting at windmills there is a (fairly well concealed) thing to think about, and it has to do with length. It doesnât bother me that much of the prose I read these days has an age measured in hours, or is evanescent electronic text, or is produced by principals rather than intermediaries. But hereâs what Iâm coming to think: in text, short form tends to drive out long form. Our novelty-seeking chimpanzee minds would rather chew through a bunch of tasty little morsels than a full balanced meal. For example, when I was just about to turn in last night, I glanced at the New Yorker magazine at the end of the sofa, got started reading George Packerâs excellent The Lesson of Tal Afar, and didnât get to bed till way past 1AM. And I learned some things about the state of play in Iraq that no succession of blog posts is gonna teach me, because the material really needs a dozen or so pages of beautifully-typeset densely-argued discourse.
This resonated with me mostly because we at RedMonk have had debates - often heated ones - on just this topic. They centered primarily around what the value of analyst reports are in today’s attention and time starved world. We, as is obvious to those who’ve been following us for a while, have mostly departed the realm of writing long, formal, multi-page reports in favor of shorter (relatively speaking), more timely analysis. We’ll still write papers on occasion - non-commisioned, of course (except for those that we ghost write) - but the vast majority of our analysis is output to our individual blogs.
That this transition coincided with a moderate but statistically significant uptick in our overall business (and at least perceived authority) is interesting, and certainly relevant, but what I’m really wondering is whether or not that’s a good thing. I believe Bray is right, and that the short form does indeed drive out the long form in many if not most cases, and that in turn is likely have profound consequences for any number of publishing businesses - ours included. Whether that’s a positive development or not is an open question.
While I personally am reluctant to invest the half hour or so required to digest papers in the 8+ page range, I’m fully aware of the fact that some subjects simply demand that sort of attention. Our COA notion, for example, is not something that I believe could be adequately delivered in the context of a blog. It’s a bit complex, and a bit nuanced, and therefore isn’t the best fit for a medium where you’re competing for attention with hundreds or maybe thousands of other simpler, easier to digest items.
I’m very much an advocate of fitting the content to the medium and the audience, and I’m curious as to what all of you think on the subject, and what the implications might be. I’ve been very skeptical of the value of analyst written white papers for some time - and not just the commissioned ones. I’ve pushed RedMonk, in fact, to become far less paper-centric because I don’t believe that to be a long term, compelling business proposition for anyone but the top 2 or 3 firms. It also is far less than suitable for our target audience of developers and architects. From my experience, vendors seem to be conflicted as well; a couple of the ghost writing projects that we’ve had in recent months have seen customers admit candidly that they didn’t expect the majority of readers to actually process anything but the conclusion and - possibly - content that’s neatly bulleted. Where that would seem to me to call into question the overall value of the white paper as an exercise and marketing tool, there remains a compulsion on the part of many vendors to churn these out. Many Analyst Relations professionals, for example, are compensated based on the number of papers they can extract out of their analysts. Is that the best way of building awareness and visibility these days, or might we see more campaigns like this one? Is it a lack of creativity on their part, or do these remain high value as marketing tools?
Without question, I have a slightly skewed perspective on this because I spend most of my time in the weeds with developers, who a.) rarely read them and b.) seldom the target audience of these things. I have problems with that approach, of course, because if you’re targeting the “IT decision maker” these days I think it’s silly not to be targeting the developer, but that’s a whole other story.
In any event, I’m very interested in hearing what some of you think of the value of analyst whitepapers and reports. Are they valuable to your efforts? Should we be doing more of them? And if their value is lessened, do you think that’s a good or a bad thing?
Popularity: 2% [?]
9 Comments
Many analyst reports tend to offer just catchy soundbites like “90% of Windows admins with no Linux experience are better at Windows”. Do vendor sponsored “Get the Facts” report have a shred of “fact”?? Who really knows - any developer/admin will tear them apart, but IT managers may not know… those analyst reports just create soundbites and somewhere in the 65 pages of garbage you find “the trick”.
As for blogs/wikis… time will tell. At the end of the day though, blogs are just people. Some have greater credibility and others don’t. Does Joe Smith posting about a T2000 home brewed test comparing it to an HP AMD system with no specs/configuration /environment details convince me of anything… not at all. What’s the credibility of the source? First, they probably got the system b/c it was sent to them for free for 60 days, they’re incented to post a review b/c the CEO of Sun promised it might be free, and Sun’s MARKETING team gets to choose who has the best reviews to keep them… Does David Smith posting about Ubuntu Linux running faster than Solaris 10 on a T2000 seem insightful/credible… yes - heck he works for Red Hat and runs the tests on Ubuntu, an interesting emerging competitor. Does this instill credibility in Joe Smith’s home cooked benchmark test? You have to consider the source.
Wikis at least have several levels of review, however, you have to consider who is reviewing. If I’m very knowledgeable of filesystems, but spend most of my time with ext3, and I even contribute my knowledge to the ext3 wiki entry, do I ever review FAT wiki entries that have possibly scewed comparisons to ext3 authored by the creators of the FAT file format?
Some amalyst reports do provide useful information though - those that include unbiased primary research tend to be best. Analyst reports like EDC developer studies tend to be more insightful - they poll and report in a more unbiased way to help the readers understand a trend of shift in the industry offering new insights. Often the ones with unique data points are of some interest. I’d pay for someone to tell me what % of IT decision makers actually read/trust/rely on Microsoft sponsored studies attacking Linux. That would be an interesting primary research data point. Others have quadrants that while biased by vendor influence are not often WAY out of place and offer at least baseline guidance. Sure they’re influenced by vendors, but I’ve never seen an quadrant putting a really poor product/vendor in the leader quadrant - some are debatable, but usually never completely off base.
The reality is that many execs don’t have the time to go through every vendors’ sales pitch, have their teams do a technical review, testing, comparisons, etc. They’re looking for that “short cut” and an analyst report’s exec summary may fill that need. Or they’re looking for a quick counterpoint to stop someone in the organization from taking a particular direction. By no means do I suggest these justify analyst reports as the best answer. I’m not convinced blogs/Wikis are there yet either though. Once you overcome the credibility factor - you can do 1/2 page on a napkin or 8 pages with graphs to make your point. Hence I read tecosystems - the author has built up credibility for me based on what I’ve read/seen over time - despite his passion for the Sox with a $120M roster salary versus my team’s $56M. Bottom line: I know that I can read a blog entry here and see an honest opinion despite some of the topics are understandably focused/related to client work that you do. It’s the credibility of the source and clarity/balance of the analysis that counts - not how long the report is.
I will also comment that the blog approach let’s users avoid going through AR to get the report… kindof how Linux penetrated the data center b/c it didn’t require going through procurement…
I think the most important thing about writing is carrying reader interest throughout the entirety of the piece.
If you’re not good enough a writer to do a long form article or paper, don’t do a long form article or paper. If the subject matter is only interesting enough for a blog post, stick to a blog post.
It’s a trade off, even if a position is nuanced and requires further exploration, if the audience wants a quick, short, and to the point explanation, they’re not going to hear the message you are trying to tell them (ask John Kerry about this). On the other hand, and especially in academic papers I see this, something simple shouldn’t generate pages and pages of exlplanation, ESPECIALLY if most of the pages aren’t in the dull “Hey, here’s the proof that this works” part that most people are going to skip over any way once it gets peer reviewed.
My main purpose in reading analysts’ reports are as a filter - as Mike says, who has the time to do a whole lot of primary search? - so you use reports (even sponsored ones, with the required grains of salt) to try and get to a short list of possibilities for closer scrutiny.
My question of Redmonk is (and I can’t recall if I’ve already asked James this question, and Redmonk Radio episode 8 (from memory) answered some of it) - from what activities does your revenue come from? I like the business model, and I’m guessing you don’t do it COMPLETELY for love, but it’s not obvious to me how you get paid.
Once again, I’m back to the same point where one size doesn’t fit all.
If I want to read something about the Iraq war, a small summary would be enough, I’m really not all that interested in the subject. On the other side, I will buy the “Solaris Internals” books as soon as I find them. You can’t get much longer in terms of number of lines than books.
Sun also had a what they called “Elevator Pitch”, a group of headlines you could tell a customer in an elevator Run. They weren’t trying to replace Whitepapers, it was a different thing, to use at a different time for different people.
Long gone are the times where the cost of printing and ddistributing information was great and it would force you to aim at the largest group. Tim Bray appreciated that long article but, he didn’t stop reading short Blog Posts on that account. the same way, most of us use Wikipedia but, We didn’t abandoned other sources of information.
Just another piece of information I think you’d like to know. As you know, I do a lot of technology choosing for our reseller portfolio and, I get to read some Gartner Magic Quadrants. I go straight for the chart and, I compare the companies Gartner evaluated with the ones on my list. All the Gartner’s analyses is trash, I stopped reading it for quite some time (the credibility of the source Mike talked about) so, all that paper, I use 2 lines
you looking to push the debate along eh stepho. this in response to anything in particular or are you just thinking out loud?
perhaps you’re making the case for the odd SHORT targeted piece?
Mike: i suppose it’s fair to shift the context of the debate from length to questions of credibility or content, and i absolutely agree that some analyst reports add a lot of value and are worth the time. i’m slightly less skeptical of blogs and wikis than you might be, however, simply because of the nature of the content. many, if not most, allow for comments and user contributions. if there are errors, then, they typically get identified and highlighted relatively quickly; unlike analyst reports which tend to be static and non-living PDFs. as for the Sox, what can i tell you - Red Sox Nation loves its team
oh, and as for the AR bit, i don’t see it so much as bypassing them as reducing their load. just my two cents.
Danno: you’re right, of course, in that some of it is giving the audience what they want. i do have some concerns as alluded to above that sometimes what the audience wants isn’t good for them, but that’s not a solvable problem.
my other question, related to that, is just how frequently users will want long form pieces these days? less than in the past, to be sure, but just how much less?
Ric: that’s much the way that i’ve consumed these reports in past.
as for questions on our business model, it’s relatively simple: we sell subscriptions mainly to IT vendors. the subscriptions are primarily consulting or advisory in nature. this of course raises questions of conflicts in interest, which is why we’re very careful to disclose our client relationships within our coverage. make sense?
Jaime: i guess it does come back to interest, then. length is dependent on what you personally have vested in a particular topic.
James: it was spurred mainly by Tim’s piece, and just thought it would be interesting to collect some opinions on the topic. not really making the case for the odd short piece; more looking for a reality check.
To add my two cents, I enjoy having longer articles that deal with conceptual issues. I think these become the seeds for lots of discussions. I also wish Redmonk did more of these.
Stephen-
This is a very interesting subject that you cover in a rather long-form blog without losing…uhm…interest
But I was talking with Andy Oram at O’Reilly and was priviledged to hear him thinking out loud about different publishing models that might be necessary to explore because the web documentation for many languages and packages seems to be quite good enough…and it happens quicker. Just look at the amount of documentation for OpenOffice on the web and you’ll see why no one can justify publishing old-school hard material on that package.
For tech books, there may be a need to publish shorter, quicker — acknowledging the short shelf-life of MOST relevant content. For example, just think how quickly AJAX has become so overwhelming of developer attention and how fast that affected many different areas.
Do find Andy next opportunity. O’Reilly has already started something called “Rough Cuts” I think to give interested audiences an early look in on Safari content as it’s written. I hope I’m not speaking out of turn about that, but it suffices to say you’re on to a hot and highly relevant topic.
Post a Comment