I began to think what a deed I’d done.
I grabbed my hat and I began to run.
I made a god run but I ran too slow;
They overtook me down in Jericho
- IN SEARCH OF LITTLE SADIE

Lordy, has it been a year already? Sure has. The 2010 Gartner Magic Quadrant for WCM is out. You can get the report here courtesy of our friends at SiteCore. As usual it is worth a read, but here is the juicy bit:

I’ve marked the guys that have improved a reasonable amount with a green line, indicating where they’ve moved to since 2009. No-one has really slipped, although a few have vanished. EMC have given up on WCM and are partnering with Fatwire instead. Vignette and Nstein are also now part of the Open Text dot. Expect to see Day replaced by Adobe on here in 2011.

They’ve stuck with the same Big Three (Oracle, Automony/Interwoven and Open Text) in the lead as last time. Two other Big Guys – Microsoft and IBM – are inching closer to the Leader Quadrant. It does seem that to be near the top of the “ability to execute” axis, you need to be a massive company and have technology that is at least ten years old. I ranted about this last year, and the same thoughts apply. I should point out that this dimension is defined as “how well a vendor sells and supports its WCM products and services“, not on the success of implementations or happiness of customers. If you want to get the products with the most marketing dollars behind them, this is the axis for you.

The Open Text logic still confounds me. Here is how I see it. In 2009, Open Text was one of the three leaders, based on what I can only assume was The Product Formerly Known As RedDot. Vignette and Nstein were lingering in the shitty quadrant (VIGN on the border, admittedly). So my only conclusion is that RedDot was the favoured product in the eyes of Gartner. However, my spider senses (and OTEX staff layoffs) tell me RedDot is on its way out and the Vignette WCM product is the Chosen One. So I’d have expected the Gartner folk to move OTEX further into the danger zone, but the uncertainly and product direction have actually given them a boost.

The tussle between the younger upstarts is as close as ever. The Java vendors (FatWire and Day) have gained slightly on the .NET ones (SiteCore, Ektron). The Java/.NET hybrid, SDL, keeps its nose in front. I think we’ll see bigger gaps in 2011.

Last year, I noted that poor EPiServer had got a bit of a raw deal. That’s been fixed. I’ve always felt they should be sitting right next to SiteCore on this thing. And CoreMedia also got a big bonus. Alterian got a little boost, but they’re still in the quadrant of despair.

There are two new vendors on there, Atex and Dynamicweb. I’ve heard of the latter but never seen them. And only heard of Atex when they aquired Polopoly as few years ago. Never seen their product either, so not comments here.

Still no Open Source vendors on here, for the same revenue related reasons as last time. I’m not going over all that again.

Most of these little yellow dots haven’t drifted very far in a year – the report is pretty similar despite the M&A activity that has kept us bloggers busy. So pretty much a repeat of last year. And, like last year, here is hoping Gartner’s lawyers don’t serve me any takedown notices.

All the tired horses in the sun
How’m I supposed to get any ridin’ done? Hmm.
- ALL THE TIRED HORSES

Where there is smoke, there is fire. In this case, it’ll be a shitstorm of a fire that’ll consume everything useful in it’s path. A bit like a Scorched Earth Campaign of Content Management.

Of course I’m talking about the Autonomy/Open Text speculation here (@hakana), here (@ldallasBMOC) and here (@piewords). In truth, the few rumours are, according to the crowds, highly unlikely to have any substance so this whole post is a waste of time. Apart from photos of CEOs in bed with hookers, no-one in Twitterville can produce a single good reason for it that I can swallow. But if this post even slightly reduces the miniscule chance of this joke of a deal materialising, it’s time well spent.

The whole is less than the sum of the parts

Onward. It’s pretty clear there is no way that this deal could make a new sale more likely. The number of different CMS and search products The Firm would have warrants the invention of a new Collective Noun. How about a Gaggle of Products? Or Confusion of Products. Or Mindfuck of Products? Should some poor customer go through a vendor selection exercise and pick Opentonomytext, they’d need to go through another one to pick the product. Ain’t gonna happen.

And the poor existing customers. Following the Open Text/Vignette deal, many poor customers are still wondering whether or when their product will be discontinued. They’re playing Russian Roulette with about 2 bullets in the chamber. If this deal happens, they’ll have about 4 bullets. Some will jump ship, so the whole idea of creating a maintenance revenue cash cow doesn’t make sense either. In simple maths terms: (Autonomy Maintenance 2011) + (Open Text Maintenance 2011) < (Opentonomytext Maintenance 2011).

Spare a thought for the search engineers at Vignette. They OEM’ed Autonomy as their search for years. “Best of Breed”, they all cried. Then arch-rival Interwoven was aquired by Autonomy. “We’re not paying our arch rival cash every time we sell a product”, they  cried. “Autonomy is a piece of shit. Let’s embed the Open Text search engine.” So they did. Hopefully they didn’t delete the code, cause they may be flipping it back pretty soon.

And spare a thought for yours truly. I’ve pushed my MS Paint skills to the limit creating the Super Spliced Open Text Logo. I think the only way I could make a logo for the new beast is on a Möbius strip, and I don’t have any lying around.

Now I’m not saying that the road ahead for either company is paved with gold. But they’ve both got some good products, some great people and a fair bit of cash. If they roll up their sleeves and innovate, they might just be okay. If they keep playing Pass The Parcel with products that develop more slowly than tectonic plates, they’re toast. Wait! Hold on a second! What’s that putrid smell? Oh, look, it’s an an elephant graveyard. And elephants don’t make good software. Especially dead ones.

P.S. Remember, this isn’t actually going to happen. Surely. They’re gonna buy someone else. Answers on a postcard.

Nobody to rescue me,
Nobody would dare,
I was going down for the last time,
But by His mercy I’ve been spared
- SAVED

An interesting play by FatWire. Our knights in shining armour have heard the shrill cries of distress from the damsels stuck at the top of Tower Vignette and Tower Interwoven and have gallantly offered to migrate them away to the safety of Castle FatWire for free. The name of the package (FatWire Rescue Program) implies the damsels are in serious trouble. According to the press release, they have a lot to worry about:

This limited-time program enables organizations that are constrained by the rigidity of their current legacy WCM products, or concerned about the future direction of their current WCM vendor

I love the sneaky use of the word legacay here. Vignette and Interwoven are suddenly legacy simply because they’ve been bought by OpenText and Autonomy respectively? Looking forward to seeing a response from VIGN/OTEX and IWOV/AU.  Interwoven is the fastest growing ECM vendor and Vignette, despite recent troubles, are still fighting and releasing some cool new things. They certainly aren’t legacy in my books.

It is going to be cheap and painless to migrate. No license costs! Woot! The press release tells us that

The program enables customers of these recently acquired companies to upgrade to FatWire’s industry-leading solutions at no license cost, when they employ FatWire’s proven migration tools and services that reduce the risk and increase the speed of migration.

Oh, wait, maybe it isn’t that cheap at all. Our knight comes with strings attached. You need to use FatWire’s migration tools. These come in the form of partnerships with Vamosa and Kapow, two heavy hitters in the world of automated content migration. My first observation here is that these two, while both offering an excellent service, normally compete with each other. So I’d be interested in learning more about the way a company (or FatWire) decides which of the two products to run with. My second observation is that both of these products can come with a reasonably large price tag. I presume the model here is that FatWire will take some cut of the migration cost in return for referring customers to Vamosa or Kapow.

The other piece of the revenue pie will come from the associated implementation services, either from FatWire Professional Services or an implementation partner. As Irina mentioned earlier today, implementing a CMS costs more than buying a CMS. And what if it isn’t only a CMS in the mix. For example, you might have Vignette Portal in there too. Or one of Autonomy Interwoven’s many other products. FatWire don’t have a product suite to replace all of these components quite yet, so this offer seems to focus primarily on customers that only use the WCM product from their current vendor.

I don’t know enough about this offering to guess whether this migration will only cover the content management and migration aspects. Any delivery side “migration” will involve a significant amount of work. Maybe FatWire already have tools in place that can either statically deploy the same files generated by a baked Interwoven site, or replace the Vignette/Interwoven API with the FatWire one for fried sites.  All three products have a Java API so at least we don’t need to worry about language-level changes too.

Maybe this is a marketing stunt. I received my email from FatWire marketing as the announcement was made. But maybe it is a genuine way for customers to save time and money if they are planning to migrate from their current platform. FatWire is a solid choice (Forrester just patted them on the back in the WCM for External Sites wave) and the content migration products are industry leaders.

The Forrester Wave™: Web Content Management For External Sites, Q2 2009

If you are considering this option, I’d love to hear more from you once you know the costs involved and how the process will operate. They are many ways that this could work, and it’s all going to come out in the wash.

You will search, babe,
At any cost.
But how long, babe,
Can you search for what’s not lost?
- I’LL KEEP IT WITH MINE

Had an interesting meeting with Bruno Pereira from Autonomy/Interwoven yesterday. He gave us an eye-opening update on the future plans of the company after the merger with Autonomy. The ink on the contract finalising the merger is hardly dry, yet I am proud to present the first architecture diagram of the new, integrated product suite. Here it is:

The Not-So-Official New Autonomy/Interwoven Product Suite

Okay, so I lied. It’s just a scribble on a flip chart. Hopefully a better version of this will come up shortly. For those of you that can’t read the writing, the boxes say things like WorkSite, OD (OpenDeploy), TeamSite, LiveSite, MT (MetaTagger), MB (MediaBin), Optimost, Virage, Meridio, etalk, Zantaz and IDOL. More here and here.

I really enjoyed the meeting. I’d not met Bruno before, but his enthuasism for the company is obvious and infectious, and he clearly knows his trade. He was able to talk at many different levels, and showed us a few demos too. I was pleasantly surprised by the vision that he painted. He assured us that the knowledge from the long forgotten Verity is all safely tucked away in IDOL these days. And he could even answer my collegue’s question about custom fixed layouts in LiveSite, which was cool. We’re hoping SP1 is going to help us out a bit.

One piece of news that I was glad to hear – we no longer need to wait with bated breath for TeamSite 7. The version numbering system is going to continue to jump in little increments. So I think we’re going to be living in 6.x land for a while.

Sadly, I forgot to ask one of the questions that has been on everyone’s lips. What’s going to happen to the other CMS vendors that OEM Autonomy in their product? For example, will Vignette continue to ship Autonomy for their repository search, or will they become yet another CMS Vendor that moves to Lucune/Solr.

All this should unfold soon. Keep your eyes glued to your RSS readers …

There’s a lone soldier on the cross, smoke pourin’ out of a boxcar door,
You didn’t know it, you didn’t think it could be done, in the final end he won the wars
After losin’ every battle.
- IDIOT WIND

Part 1: The Beginning
Part 2: The Meme Spreads
Part 3: The Aftermath

Right, it is time to draw this chapter to a close. If you don’t know what this is all about, please read Part 1 and Part 2 mentioned above first. There has been a great response, and there is talk that an ECM focused meme will start soon too. This has been reported by Julian Wraith, who also did an excellent job keep track of all the responses. Thanks! Google Bertrand Delacrétaz’s Meme ID to find everything there is to find: 9c56d0fcf93175d70e1c9b9d188167cf.

For the record, the 24 vendors that responded (with links to their responses) are: Jahia, Ektron, Hippo CMS , Magnolia, EPiServer, Nuxeo, GX, Midgard, Knowledge Tree, infopark, Day, Alfresco, CoreMedia, Sitecore, OpenText, Alterian, dotCMS, Vignette, Autonomy/Interwoven and Escenic. Not yet in the table: eZ Systems, e-Spirit, Enano, Sense/Net.

Below is my summary of the results. Click the image for a larger version. Please note that everything here is extremely subjective. You’d be insane to use this as part of any vendor selection exercise!

Summary of the Scores

Notes on the scoring:

• Vendors shown with a blue background did not score themselves. So I scored them based on what I think their answers meant.
• GX gave themselves a 0 and a 1, where they meant a 1 and a 2. So their score rises from 40 to 42.

The Overwhelming Yes Questions

These questions had an average score of 2.8 and above:

• Amazingly, everyone thought their Software Does What It Says it did. That’s great news for the CMS buyer. Everyone is honest . Most don’t even try to justify this, although Ektron justify their reply with the fact that they have “more WCMS implementations than any other vendor in the marketplace”. I’d be interested to see the list/research on which this is based.
• They all claim to have a Free SDK. I wonder if Kas Thomas had a vendor in mind when he put that question onto the list?
• The No Reboot topic caused a bit of controversy, with three vendors claiming it isn’t important. I disagree with this. A reboot doesn’t cause any downtime in a load balanced environment, but I feel it is architecturally wrong for the CMS to go so close to the Operating System to warrant one.
• Alfresco were the only ones to admit that they need their technical presales engineers to help the Sales Guys. Maybe the other vendors included these techies as part of the sales team so gave themselves full marks.
• Again, Alfresco were the only ones that confessed their Sample Site wasn’t great. I’ve seen a lot of these sample sites, and some of them are really quite dire. I think a few more vendors should have given themselves a “Sort Of” here. But I guess the question did not ask for a Good sample site, so they are technically honest. They all have a Tutorial, most of which are good.
• Three vendors admitted that they didn’t have a Full Installer. dotCMS lost some points here, but at least it is on their roadmap. I do think that the Hippo were a bit hard on themselves. I prefer a standard EAR/WAR deployment to an install Wizard, especially in a large, clustered environment. A One Click Update question would have brought some interesting answers as upgrades are often much hard than installs.

The Differentiator Questions

These questions had an average score below 2.8:

• About half the vendors struggled on the No English challenge. The smaller US based vendors normally do worse here as many of their clients are single language. The European based vendors live and breath multi-language every day.
• Interwoven is the only vendor that does not offer a Download. Six others offer one with conditions attached.
• Most vendors eat their own Dogfood, apart from SiteCore, who drink their own pre-release champagne. Quite a few aren’t on the latest version though. Escenic gave themselves a 1 for this, while many other vendors gave themselves a 2. And I’m taking bets on when Vignette’s site is going to be running V7.6. KnowledgeTree gave themselves a very kind 3 seeing as they don’t do WCM. But their score does prove that most of the questions are noting to do with WCM and could apply to almost any software vendor.
• Escenic were the only ones that admitted their Price List is able to “adapt to a large variety of customers”. I believe that all the major vendors do this, and that the price can vary enormously. Maybe the answers refer to list price as opposed to the actual golf-course price, but I think Escenic showed honesty here that some of the others could have done. Vignette did also admit their model is complex. Half claim a 5-year old could understand it. But I hope this five year old can also drink a lot in the pub and negotiate a good discount or he is going to get screwed.
• Surprisingly, the question with the lowest average score was the Raise Issues From Product one. All vendors have an issue logging system, so this would be extremely simple to implement. It could just be a link from the admin screens to the support screens.
• The most controversial question was probably the All Help Files And Documentation are Part of the Install. Now I’m going to disagree with Kas on this and side with some of the vendors. I prefer online documentation to local installation for a number of reasons: It is kept up to date and continuously enhanced, it can include user submitted contributions, it can be powered by an advanced search engine, it can contain offsite links, and I don’t like having extra items installed on the servers. I do like to have a local of the SDK, but this should be embedded in the IDE and be an optional extra download as part of the SDK. And if people do still work when they are offline, the ability to download all the documentation would be a bonus.

The Jon Vendor Meme Awards 2009

Any great competition needs an awards ceremony. So, in the spirit in which this whole contest was conducted, I’m honoured to be able to announce:

• The Jon Award For Transparency goes to Escenic. Sure they came stone last, but I trust the guys.
• The Jon Award For Agility Above And Beyond Expectations goes to Vignette. I really didn’t expect them to respond, yet they were the first ECM player to do so.
• The Jon Award For The Best Product Name goes to Hippo CMS. Of course.
• The Jon Award For Anti-Cheating goes to Nuxeo, who miscounted their score on the low side. I’ve added a point to their reported score.

And thanks once again to Kas Thomas for his “A reality checklist for vendors“, and Day for starting this party, giving CMS geeks like me something to smile about. It’s been real.

UPDATE: 28 March 2009 – Added KnowledgeTree and Ektron.

UPDATE: 03 April 2009 – eZ Systems have responded. I’ll update the chart when I have more time.

UPDATE: 08 April 2009 – e-Spirit set themselves up for a fall by giving themselves a perfect score. Anyone know enough to see if they’re being cheeky? Not a vendor I know much about sadly. 45/45 seems a bold claim to make. They didn’t publish any contact details for question 15. I notice @espirit_news joined Twiter 15 minutes ago. Just in time. Well done, e-Spirit! Thanks for playing.

UPDATE: 09 April 2009 – Adding Enano (seeing Julian included them in his list) and Sense/Net.

I ain’t lookin’ to compete with you,
Beat or cheat or mistreat you,
Simplify you, classify you,
Deny, defy or crucify you.
All I really want to do
Is, baby, be friends with you.
- ALL I REALLY WANT TO DO

Part 1: The Beginning
Part 2: The Meme Spreads
Part 3: The Aftermath

It was the morning after the night before, and the CMS Vendor Meme started by Day Software began to spread. Two further vendors (Magnolia and Alfresco) were infected on the first day, but the days that followed saw the meme spreading like the Undead. So far, we have 18 Vendor Zombies. What follows is a day by day account of the unfolding events.

dotCMS have good designers

Big news. Vignette have responded using a FaceBook note. I gotta be honest, I didn’t think they would. Hats off to them, I say. They started with an intro explaining how Kas’ checklist didn’t really apply to them, only to ” stand-alone, SMB-focused software targeted at companies” that are “managed by a very small IT team”. They also find “that the Enterprise customers we serve don’t typically let the needs of IT drive their Web experience decisions”. So what? Why does that mean that the checklist doesn’t apply? I would have left a comment, but sadly no comments allowed. [UPDATE: Mea Culpa. You have to be a Fan of Vignette to comment on their page. But you can.]

They didn’t give themselves a score, but gave a YES for most questions. The occassional NO was well justified. The one SORT OF that made me chuckle was the “running on latest version of the software”. They’re “not on the latest point release”. Well, have fun upgrading that. It’s only a point release after all. Shouldn’t take long at all. Just a click, I’d think …

20 March 2009: Nuxeo enter the fray with a 40/45. They haven’t tagged anyone since they think there isn’t “any serious player that hasn’t been tagged already” . And they seem to be my kind of company. They’ll trade SDK’s for beer. They’re threatening to spread the meme to DM and Collaboration too.

Sweet! Another big player is up for a bit of fun. OpenText have entered, but not given themselves a score. Like Vignette, they start with a well written but very corporate intro explaining how the rules of the game don’t really apply to them. Nevertheless, they kindly agree to play anyway as “our customers expect nothing less than a transparent, strategic vision from Open Text”. Impressive stuff. An extremely slippery dodge on Question 14 – one price sheet. They also manage to squeeze in a few sales pitches (question 9 and 10). The RedDot name didn’t appear once. The brand is truly gone it seems. All in all, nice one OpenText.

Hot on their heels comes EPiServer, who score themselves highly with 42/45. Right up there at the top. Now, I am a skeptical little man and that sounds on the high side. But I know EPiServer extremely well (believe it or not, I’m a certified developer who isn’t allowed to code) and I don’t think I can’t fault their logic. They’re certainly very accurate about their answer to the language question. If anything, when you install it as English you might see the odd bit of Swedish. And maybe I’d lower the 2 to a 1 on the licensing model as the definition of a site, as they confess, is confusing as hell. But a big thanks to EPiServer for entering. They haven’t tagged anyone yet.

SiteCore in next. They gave themselves a modest 40/45. A very nice response though. My favourite was the response to the Dog Food challenge – not only do they use pre-release versions on their site, but they call it Champagne instead of Dog Food. Also like their answer to the documentation question. I’m starting to think that the Documentation question doesn’t belong on the checklist. Online documentation is probably more useful than locally installed documentation. Thanks for playing, SiteCore. But I’m not going to link to your ugly red “star image” cause it is 500 KB big.

Autonomy/Intervowen and SDL Tridion – come out, come out, wherever you are …

21/22 March 2009: It’s pretty quiet over the weekend. Almost too quiet. Then, suddenly, another big player posts their response. Thanks to Tom Wentworth from Autonomy/Interwoven. It’s absolutely brilliant to have all the major vendors getting involved in this. Really good for the CMS Community, I think. Like the other Enterprise Vendors, they haven’t given themselves a score. I think if they were to score themselves, it would be pretty low. However, all their responses are sensible and highlight the fact that the ECM players considers themselves very different to the smaller companies.

So, it seems that only one large vendor that has been tagged who is staying out of this. I’m not going to name the chickens again, but I wil say that an anagram of their name is “sordid lint”. C’mon, people, air that dirty laundry.

23 March 2009: Things appear to be getting quieter now that most of the main players are in. Doesn’t look like the meme is going to infect non-CMS vendors. A new entry today comes from Alterian (who own both MediaSurface and Immediacy). I liked the tone of their response. Very collaborative, very honest, and even congratulating some of the vendors on their response to other questions. They, like others on Twitter, question the validity of the reboot on install challenge, about which I’ll say more when everyone is in.

They don’t give themselves a score, but we’ll forgive them as they would probably have to score their two products independently. This would confuse the league tables horribly. I think the tone of their response proves how wrong I was about how this was going to play out. It isn’t a Deathmatch at all. But “CMS Celebrity Deathmatch” is a more catchy title than “CMS Collaborative Love In”.

“Some think the noble hippopotamus a somewhat pointless rhinoceros“. Not these guys, though. Hippo CMS come next, tagging Open Source Vendors OpenCMS, eZ Publish, Joomla!, TYPO3, Plone and Drupal. If this meme spreads into the Open Source land, all hell could break loose. Maybe someone will tag WordPress? Anyway, Hippo join Jahia at the top with a whopping 43/45. And where they deducted points from themselves, you can tell they feel they have the moral highground. Who needs an installer when you can deploy WARS and EARS? And surely editors speak more than one language? Did you know that hippos kill more people in Africa than any other large animal?

Another great side effect of this meme is that we are finding CMS Gurus on Twitter to follow. They’re all crawling out of the woodwork. Bill Trippe is compiling a list. He said he’d post a link here in a comment when it is ready. [UPDATE: Bill has posted the link below, but I got this wrong. It is a list of the actual vendor accounts, not individuals].

Part 1: The Beginning
Part 2: The Meme Spreads
Part 3: The Aftermath