Tip of the Iceberg

1 - Getting people to treat Domino "seriously" as an application development platform is going to take more than merely changing some of the licensing requirements for external authenticated users.

For background, I got into professional IT between 1994 and 2002 doing Lotus Notes and Domino work in Canada and the US, after a spell of Pascal, Assembler and C work in high school and college. I haven't done any professional Domino/Notes work since 5.x, and haven't done any hacking with it at all since the 6.x days, so if the platform has moved on and my opinions don't reflect reality, apologies. It's not deliberate; just the perspective of a Notes alumnus who's left the platform behind but still reads PL.

While I left Lotus behind professionally for the purely mercenary reason that it became impossible to get a decent-paying Lotus job in Atlanta (or the Southeast), I left Lotus behind personally because of its failing as an Application Development platform for a number of reasons. Since then I've done some work in Java and C# on the web side, the Windows client platform and back-end services and gained an appreciation of what other platforms, not just Domino, could do.

In 2001, the Lotus Business Partner I worked for in Atlanta decided to get out of the Lotus/IBM business and I was left unemployed for several months with a very pregnant wife and a dedicated, unrestricted 144K internet connection with a static IP and a Windows NT box running Domino. To keep my skills sharp, I created a web site for a flight simulation community called a "virtual_airline" where folks pretend to fly 747s in MS Flight Simulator and needed somewhere to log their flights and hang out and chat.

Domino was a great tool for the job. As folks advertise, its RAD abilities were great and within approximately a week I was able to create an attractive site that was better than any of the others out there, and with Domino's integrated services I could offer webmail, LDAP services to external parties and all of this within a few weeks. I expected we might get 100 folks at most, and after two years we had several hundred. It was amazing.

At this point, the "dark side" of Notes and Domino reared its head. Just like I saw when I did Notes consulting work, we had an application that was put together very quickly and was successful, but the more successful we got, the more Domino got in our way rather than enabled us. More advanced features required Domino "hacks" where we had to code around the platform, and that made things harder to understand or maintain. Add to that the fact that I didn't have a legitimate Domino license for the Server, Designer or the fact that I had several hundred user accounts in a cascaded NAB, and I simply couldn't stay on this platform.

In late 2004, having not worked with Domino professionally for over a year, I decided to rewrite the site entirely in Java using OSS technologies (such as Eclipse and Tomcat) and mySQL. The first four months were nasty because I needed to create a large number of JSP tag libraries to make creating forms, fields and other Domino concepts in a J2EE world just as quick, but once this was done, development sped up and the same "eyes lighting up" moment that others have when they first see Domino were occurring to me. Transactions? Joins? Relational integrity? Data type validation? Oh yeah. So many things that were hard with Domino became really, really easy to implement. I won't diminish the up-front investment I needed to make. It was not trivial, but at the same time the return I've gotten on it since then has been well worth the price.

I wrote my last bit of Domino code in early 2005, the new site went live in September of that year and I haven't looked back seriously since. But my experience leads me to a question you want to ask yourselves - how many folks in the Yellowverse have become completely fluent in one or more platforms other than Domino and still love it? How many have done open source work rather than Sharepoint? It's nice to denigrate MS but the real threat to Lotus as an app dev platform isn't whatever MS is coming up with - it's the open source technologies such as PHP, Java, Ruby and others.

So with that said, here's what I think Domino needs to become relevant as a development platform. I'll repeat my caveat that I haven't written code in three versions and five years, so if the platform's met these goals, good for you.

Make licensing irrelevant. Nathan, you're right, and others who have posted recently are too. If the number of users I have dramatically affects my licensing cost, Domino has lost. If I even consider licensing as a significant part of my app design, Domino has lost. I shouldn't need to worry about CALs, PVUs, processor cores, instances or any other details. I just want to write great apps that do great things. Now I expect someone with an IBM enterprise license doesn't need to worry, but for everyone else? They just move to PHP, Java or something where spooling up another instance or adding cores has no licensing implications whatsoever.

Make NSF a real database. Yes, it's not relational and will never be. But at least support transactions (how many times have you needed to update multiple documents and want it to be atomic?) and give it some basic data integrity protections. The lack of a rigid schema is a good thing, but being able to enforce a schema is too. I've lost track of how many times I had great apps in QC bomb in production because I could make no guarantees about the contents, data type or presence of a field. This is something that NoSQL needs to deal with as well, and they're moving on it. While we're on the subject, look at what the Dynamo clones are doing with requiring reads/writes to be made to multiple instances before they return success - this is yet another case where someone else is taking what Notes did a decade and a half ago and building on it while the Yellowverse is busy patting themselves on the back.

Support version control in big teams. I'm working on a project with a dozen devs and designers, and using Subversion it's trivial to get the contents of prior versions, find out who made a change (and shoot them ), resolve merge conflicts all from the IDE. DXL appears to be a good start, but Domino needs to support SVN, Hg, git or other popular packages. Make an API to do this (I imagine with Eclipse it isn't hard) and ensure that your data storage structure (DXL? Something else?) makes browsing changes and merging easy.

Let me control the HTML. I imagine XPages fixes this, but if it doesn't 100%, you need to fix this forthwith. And ditch Richtext. If I need Ben L's voodoo to control what I'm doing with the core of Notes.

No 64K limits. Please tell me that in a world where I can deal with terabytes of data in RAM there isn't a single 32K or 64K limit left in Notes. Please.

Decide what scale of apps you want to do. Like I said earlier, I saw tons of apps in my career that became valuable with 80 hours of work but after a year or two became an unmanageable mutation of hacks and work-arounds as the business saw the value and started asking for new features that weren't necessarily as easy to implement. It's no shame to decide that certain kinds of apps are "too big" or "too small", but Notes doesn't solve all problems. And if we discover than Notes doesn't easily solve problems we're pitching it to do, then find out what's wrong. And make it go away. I've identified a few areas that I think fall into that category.

Also, compare DDE to Eclipse or VS2010 and make sure that everything that devs are used to in those platforms are available in Domino. Make a change to a library API signature? I want to know what fails to compile in real time. I want real-time context-sensitive help, and be able to find every reference to a method, field, hide-when formula (OMG!) library or class with a single right-click. Because I can now.

Ditch the client. If you want to be relevant as an App Dev platform, forget the Notes client. The "client" is a web browser, either on a PC or a mobile device. Proprietary clients just diminish your credibility with developers doing something interesting outside of boring, internal corporate apps. And corporate apps suck. Why? Because they people who use them don't get to pick them. End of story.

Pick a language. It amazes me that after almost a decade and a half since the first Java API for Notes, it still doesn't do everything that the LS or @Formula APIs do. XPages appear to have added JavaScript, for four languages. Figure out what language you want to support, and make a developer able to write every single line of code in that language alone.

Pick a developer. Figure out what level of developer you want to target, or which developer your feature set targets. Either way, but decide. It's a much harder problem to target the "power user" along side a top-notch software architect who wants the same features that advanced languages and IDEs provide. It's telling that Domino devs are complaining about how hard Java is when most advanced developers dismiss Java in favor of more advanced or flexible languages. I fear that most of the advanced devs left Domino and Notes about seven years ago.

You can probably make a platform that targets both communities, but your job has just become three times harder. Figure out what level of skill you are aiming at. I fear that the existing dev community has made the decision for you, and because of that Domino's potential is intrinsically limited.

Bring me back. Okay, maybe I'm a lost cause. But if Domino wants to be relevant as an app dev platform, it needs to provide a case for people like me who've moved on. I write every line of code on a hobbyist site with 2,500 users (deltava.org) and I am also the lead architect on a site with approximately 10,000 times as many users (weather.com). Can Domino provide a compelling proposition for either of these sites to make me want to write new code on them? If you can answer that question, there's hope. If you can't, then you're not going to get the Facebooks of the world (as one poster has hoped) or the next generation of devs. They're too busy wanting to create (and in some cases, actually creating) the next Facebook.

Stop talking and thinking about e-mail. Years ago, one of the top Google results for my name was a response I made on the LDD forums where I said that Notes didn't do anything particularly well anymore. It's not designed as an e-mail client (which hurts it there) but the focus on e-mail has diverted IBM away from making it a better dev client. That's probably not a realistic goal of mine, but at some point IBM needs to write a proper mail/cal package rather than bastardizing Notes into something it was never designed to do.

That got a fair bit off my chest. Hopefully some of it is still relevant. I really enjoyed doing stuff with Notes/Domino, but I grew and the platform didn't. Admittedly that was probably the worst time to be a Notes/Domino developer, but even since then the state of the art has moved on, and I'm not sure IBM has managed to keep pace.

Cheers!

Luke

2 - Great idea Nathan. It would REALLY make our Domino world a better SALE. IBM are you reading this. Lets do this.

3 - Luke, you are behind the times my good man. Xpages addresses most of these issues, and where it doesn't yet, work is underway. The biggest leftover item is svn, and there is active work in this area.

The problem, as acutely identified over the last two weeks, is not engineering. It's marketing and sales. The code pretty much does what it needs to do. IBM simply doesn't think of the product in the right terms.

4 - I can;t begin to count the number of Domino deployments we could have sold in years past were it not for the licensing terms. I love your proposal. I hope the powers that be at IBM read it.

5 - It's a great idea. Would really open the doors to more with the Domino Server.

6 -
Add my vote for this approach as the SMB's in my area would benefit greatly.

Robert

7 - I think that anything that stands to increase Notes usage, while not reducing Notes revenue to IBM by a significant amount, has to be a good thing in the long run for both IBM and the wider Notes ecosystem.

Nathan's idea of allowing authenticated access to non-employees of a company that owns Notes licenses would seem to offer the potential for just such a situation, where
- IBM may lose a small amount of revenue on the currently (small number of!) licensed extranet users, but
- this could/should be offset by a) more people staying with the Notes platform now they can do this, and b) net new usage of Notes as it becomes a viable candidate for such usage.

In addition, it looks to me like Notes has low market share in very small organisations (say less than 100 people). If this is indeed the case then what would IBM lose by giving away free licenses to companies of this size? The potential benefits of offering free licenses to micro-companies could/should include:
- An increase in Notes usage in this sector
- Some oxygen for current Notes BPs who could target this sector with lower cost apps
- Attract interest from younger developers / college students
- Increased visibility for Notes as the low cost Swiss army knife for business apps., and helping it to compete in the larger orgs.

8 - @40, I think you are correct. The express license is a bit recursive, you need a license for everyone in the company but then (ii) stipulates for each "Authorized Users":

"(ii) Licensee must acquire one PoE for Domino Collaboration Express for each Authorized User who accesses the Program"

Still I think you are correct, but IBM needs to do a far better job indicating this.

I've tried to distill this down some { Link }

9 - @31 - Why is this a "loss leader?" The only possible sacrificed revenue is Utility Server licenses in extranet applications. How many of those are there out there, really? I have no idea, but I'm highly skeptical that it's a big number.

I'm not saying make it free. I'm make it free for users who aren't on the payroll of the providing company. That's a huge difference. The BPs, Siemens, India State Banks and NTTs of the world still have to license their internal users. Even Elguji has to license internal users. But if BP wants to buy Ideajam and put up a public site for suggestions on how to clean up the Gulf of Mexico, they don't have to pay IBM 5 times as much as Bruce to do it. They can spend that $50K on containment booms instead.

10 - @42, 43, There is potential to combat the "toy grade" problem buy simply sticking on the word enterprise and selling for huge markup. So you could have an express/lite/free version that is missing some little used feature or requires a little more effort to get hold of and give it away. Slap on the Enterprise sticker and big bucks price sticker and away we go. How about some integration to existing big bucks enterprise components? Surely that would do the trick?

11 - Not to prolong the licensing disucssion, but from the license:

"An Authorized User is a unit of measure by which the Program can be licensed. An Authorized User is a unique individual inside or outside of Licensee's Enterprise with a specific identity that is validated when accessing a IBM Lotus Domino server. The Authorized User's unique identity could be defined by his or her Notes ID/password ("PW") combination, his or her IBM Lotus Domino HTTP Name/PW combination, or any other third party authentication source that defines the Authorized User's unique identity."

That last line sounds to me like "if you employ any method to authenticate a user, they need a license" (which could be interpreted to mean, for example, using a previously assigned token in an anonymous HTTP session to authenticate a REST transaction ... which can be sweepingly problematic!)


@Darren, the license actually reads (for express):

BM Lotus Domino Collaboration Express

"Where the Program is IBM Lotus Domino Collaboration Express, (i) the Program is licensed on a per user basis for use only by Licensee's employees and independent contractors, and those of a Related Company"

No anonymous access, because no access to anyone outside the company...and anyone inside the company that accesses the "application", needs a license

12 - I do not want to drive this in a licensing discussion. I am extremely busy in Micro SMB so I fight a lot with Express licensing and its hurdles. If it changed then it must have happened recently.

"If you do not have a Lotus Domino Collaboration Express per user license, you are not entitled to access the Lotus Domino server, even if the you have a separately acquired client access license or only want to use applications anonymously from a Web browser."

That is from the FAQ from here
{ Link }

Anonymous HTTP access is only free for the Non-Express offerings. If you want Anonymous HTTP you either need Utility Express (which also provides authenticated access) or a single Enterprise server that would provide Anonymous Access only.
Both options combined with PVU nonsense make it extremely difficult to bring Domino to the web for small companies.

There is some kind of a hack for Foundations. I call it a hack because Foundations afaik normally runs Apache on Port 80 but there are valid workarounds around that (and Bob Baehr is selling a product that will make it happen).

The more you deal with Express licensing the more you will find it extremely unattractive and if you have more than 100 seats CEO seems to be the way to go (my opinion).

If you want to push XPages development there has to be something that is easy to understand and reasonable cheap. If it is not free then it should at least be in the range of Microsofts Web Server 2008 (list price $469).
And to have a complete package IBM should at least tolerate if not support a free Linux distribution (like Ubuntu server, CentOS). I run Domino on Debian since R5 and it really became a no brainer through the years for SMB setups.

Of course Anonymous HTTP access would not be enough for most XPages applications.
I would also support a license that requires becoming a customer reference. I think Microsoft has this for for one of their Start Up programs. I always find it great if you evaluate a product and can see it in action.

13 - @46 Ed, can we also get a clarification as to "any other third party authentication source that defines the Authorized User's unique identity" ... I am assuming the intent is an LDAP directory, etc. However, my question on things like an authentication "token" used to authenticate an external "app" via a REST transaction or the like ... it may be tied to a single user who created the authentication, but the published data may be used by anonymous sources ... so its a hybrid ... I am hoping we don't have to license each potential "user" of a REST transaction...

14 - I agree totally on the licensing price issue. However, let me take a contrarian view for a moment.

At work, the single biggest threat to our Notes app environment isn't MS/Sharepoint, it's Oracle Business suite. We build stuff and all stuff has a PN and every time you create a non-Oracle pile-o-data relating to PNs, you add inefficiencies to our operation and everything has PNs associated with it (we use Notes apps to help run our production floor and quality department).

If I were to do a complete upgrade of our environment, buy CALs for web facing dbs, new versions, etc, I'd be looking at way less than $100K. When we budget, those numbers get a big thumbs down.

When we look to migrate Notes functionality to Oracle (for instance our RMA handling system), the STARTING point for Oracle is $250 K and that's before you pay the consultants big $$ to customize the system so it actually works for us.

Want to push stuff to the web? Another quarter mill...

Nobody blinks an eye.

There's more going on here than cost. For some reason, Oracle is viewed as 'enterprise grade' and worth the big dollars. Notes is some kind of hobby tool so $100K is deemed unreasonable.

Nothing I can say or show changes this equation.

It's not just Oracle. About 15 years back, the company I worked for (BIG pharma) refused to consider Domino.doc for distributed doc management even though they were one of the largest Notes customers in the world. Instead, they went with Documentum, again because it was Enterprise Grade and Notes wasnt'. Documentum was a quarter mill for each site in the door before customization.

I have no freaking clue how you get this kind of mind share but I know it exists and just crying 'lower the price' isn't necessarily going to fix the problem.

15 - @44 - But they do that today. Utility server, which is exactly the same kind of Domino box you run on Foundations, provides unlimited access for any users, anonymous or authenticated, no matter where they're employed. Of course, it also has a 6 figure pricetag in any reasonable implementation.

I've only seen it in production once in my career, though.

16 - Maybe this is the issue:

"Where the Program is IBM Lotus Domino Collaboration Express, (i) the Program is licensed on a per user basis for use only by Licensee's employees and independent contractors, and those of a Related Company (collectively, "Licensee's company"); (ii) Licensee must acquire one PoE for Domino Collaboration Express for each Authorized User who accesses the Program; and (iii) the Program use is limited to companies of no more than one thousand (1,000) employees and/or independent contractors; therefore, Licensee's company may not acquire more than a total of one thousand (1,000) PoEs of Domino Collaboration Express or Domino Messaging Express combined."

I think the line "(i) the Program is licensed on a per user basis for use only by Licensee's employees and independent contractors" is what Henning is referring to (or more correctly the FAQ) in the *actual* license, and could be taken as denying anonymous access in an Express environment.

Jeez, IBM I thought you'd simplified the license ;)

License : { Link }

Still even with all that said, the term "Authorized User" in the license seems to indicate an "authenticated user".

17 - Well the FAQ and the license read different then. There is nothing in the license (and that is what a user accepts when installing the software) that supports the notion that anonymous access requires a license (see my earlier post).

Time for a blog post me thinks!

18 - IBM certainly needs to do something along these lines. One thing I'm sure they would baulk at would be allowing external users to use mail and calendar features. But accepting that there are corner cases that would have to be thought through, this is an excellent suggestion.

I'm sure there are plenty of other possibilities in the realm of radically changing the licensing structure, lets see if IBM are willing to make any moves in this direction.

19 - @42 - my boss makes exactly the same argument with me: that the price dictates the perception of quality, and therefore price reductions result in the perception of a "toy grade" product.

I don't have a good answer to this. Corporate conspicuous consumption exists. It's hard to fight. But I do know that if you could point to competitors, suppliers or customers that used Domino, it would help your case, and that your odds of doing so increase with a reduction in effective price. You can't justify the purchase of a Mercedes with "Honda sucks" when your mom, dad, neighbor, lawyer, accountant, uncle and spouse all drive Hondas. The argument just doesn't hold.

20 - @35 - Okay, I see what you're talking about there. This is news to me, and I stand corrected.

That is outrageous. If I get a Domino Collaboration Express server, I can't put a damn BLOG on it!??!?! That's the dumbest thing I've heard in a long time. WOW.

21 - During a license audit IBM's interpretation of what required a CAL varied from that of a consultant who set up a website for my friend's company. IBM conducted their audit remotely and e-mailed my friend's manager a bill for a little over $100,000. After two months of wrangling they created an LLC, licensed one employee, bought Utility Server Express, and saved themselves a boatload of cash. IBM was perfectly fine with it.

Based on my experience, doing away with this onerous licensing would help IBM a lot more than it could possibly hurt. This company has a lot of IBM's portfolio deployed and now they only use an IBM platform as a last resort, so IBM is losing not only Domino revenue, but Tivoli, DB2, Rational and Websphere.

22 - " You all want the IBM salesforce to be interested in Domino, but want to (seemingly) kill Utility Server's price point."

Yes and yes.

"Guess what -- millions of dollars at stake here." Okay. You're the only person around that actually knows how many millions. You made Designer free and that was millions of dollars at stake.

If Utility server is a $10 million business, then I there's a debate to be had. If it's a $100 million business, then there's no debate, you should keep on selling it, and just come up with better Application Specific Licensing.

23 - @46/@48 - Could a "community" version of Domino Utility serve both needs? Could there be a domino build that has zero support for the notes user mailbox, no IBM support outside of "community" support like forums, etc, that fits the bill ... i know there are more specifics to be thought about there, but its more of the MySQL model ... IBM can still sell and support expensive large scale Utility servers as they do now, but the door is opened for lower end implementations to get up and running on Utility, to foster applications built on the platform and eventually turn into larger Utility server sales for IBM ?

Honestly for an ISV trying to do hosted apps, the incremental cost for Utility Express isn't horrendous at the $28(or what ever it is per pvu) for application models that generate revenue or have proven success, but the big barrier is getting to that point ... its very easy for a start up to base its app structure on things like lamp mysql ruby etc etc because the barrier to entry is so low, actually non-existant, you can download, install, develop and deploy ... with the free designer, you can't get past develop (granted there's Amazon EC2 but its not the same as having a box right next to you to work with) so it precludes the selection of the technology...

24 - and just as a follow up thought ... if the idea of an IBM branded XPages platform is intriguing to IBM, and is somehting something to pursue, could that be the perfect opportunity to use the community model ? BTW, I LOVE the idea of an stripped down XPages platform that is Notes compatible.

25 - #42/43 hit on a key point...this licensing topic is one I'll pick up on my blog later this week. You all want the IBM salesforce to be interested in Domino, but want to (seemingly) kill Utility Server's price point. Guess what -- millions of dollars at stake here. Will be a good discussion topic.

Anonymous usage is always allowed through the web for any Domino server. Authenticated access defines an authorized user defines the need for a license.

26 - Thanks for your kind words and replies. I suspect I've been gone for too long to give any semblance of an informed opinion on the relative merits of XPages and whether it addresses all of the issues I've mentioned. But with that, I don't think all of the development concerns I've pointed are addressed by XPages. Does NSF support transactions? Is the LS API equal to Java 100%? Can I write field translation and field validation formulae entirely in a language other than @formula, and be 100% searchable? Is recycle() dead? I hope it is.

But that's a quibble. Think of the baseball rule - the tie goes to the runner. Even if all of my concerns have been met, I still need to have a tangible improvement over the platform I'm currently developing with in order to change, assuming the costs are equal. If I have to pay a license fee, the hurdle just went up.

It's no different than the argument made on e-mail migrations. There's absolutely no business value in moving from Notes to platform X if at the end of the day your feature set and capabilities remain the same. You need to have a clear advantage.

Unlike ten years ago, the landscape is one where almost every opportunity already has a web strategy and platform. When I was in the industry, an Internet presence was something new and Domino had a natural competitive advantage because it supported the web and we weren't migrating a client away from an existing technology. If someone has an ASP, JSP or PHP-based web solution, how will Domino 8.5 wean them off?

This is a genuine question, because that's the question you, IBM marketing and everyone else needs to answer. If there's a good answer, Domino becomes relevant again. As Tim mentions, XPages may have closed the gap. I hope so. But if I'm going to get back into the world of commercial licenses, it's not enough to be the same, XPages needs to be better than what I ahve. NSF needs to be better than what I have. Not just as good.

It's not enough that DDE is free, although that is an overdue development that hopefully will keep Domino in existing shops and encourage some more development. That is, however, a defensive move - it keeps Domino where it is but you need the entire stack to be free below a certain scale to attract sites. Think of my hobbyist site - we have 2,000+ members and around a $2,500 budget, most of which is consumed by hosting. Does Domino fit into our budget?

Then think of my day job - we have around 2.5 million registered users. Does Domino fit into our budget? Probably - but how do I go to our CTO and explain an additional licensing fee of $10k, $25k or $100k in order to get features no better than we have now? (And yes, 8.5.x may have closed most of the gap. To the CTO, I need something where Domino is clearly better, backed up by something quantifiable.)

The two sites I work with bracket the range of what I think Domino can do. I apologize that the questions are hard and it appears that I am not giving the benefit of the doubt, but if the platform wants to survive it's not enough to get a win or two from Exchange, or staying alive in an installed site. It needs to get some competitive wins against non-Microsoft installations. How many PHP, RoR or Tomcat sites did Domino gain seats or sites against? That's the metric you want to judge yourself against. And to be fair, how many custom VB6 apps did Notes client apps displace? That's a good metric too if you still believe in the Notes client.

At some point I should probably download 8.5.1/8.5.2 and see what it can do nowadays. I feel a little guilty being such a skeptic, but at the same time I don't think my questions are unreasonable for a CTO or enterprise architect. If Domino wants to be a relevant app dev platform it's not enough to maintain the status quo - there needs to be some areas where its features outweigh the licensing issues to steal a few wins from the OSS competition or ASP.

Cheers!

Luke

27 - Nathan - I absolutely agree with your viewpoint, however I respectfully think it misses what Sean was trying to get at.

As we're also a company with less than 1000 employees we also have the Express licensing which for us means we just pay for the seat costs and no server licensing costs (with restrictions). As soon as we go over the 1000 employee mark we are no longer entitled to Express licensing and therefore have to pay full price for our Domino environment. This is irrespective of non-employees accessing our site.

While I understand that the Express licensing is aimed at making it easier for SMEs to implement a Domino solution, unfortunately, at least here in Australia, IBM haven't followed that up with a targeted push to introduce Domino into this sector.

As someone who is involved with mergers and acquisitions I get a lot of funny looks when I say that our entire system runs on a Lotus Notes platform. I then go to great lengths to point out that 3 of our Big 4 banks use Lotus Notes/Domino (who are also our clients) and that IBM is great on technology but not on marketing, and Microsoft is the other way around (only a few get that not-so-subtle dig ).

We're currently toying with the idea of taking our property appraisal application to the US, however a big question mark to me is whether we do it now as a Lotus Notes client-based app, or whether we spend the extra time and money to webify it with XPages (or RoR). Of course the decision is more complex than that, but it's certainly near the top of my list of questions/concerns.

Sorry for babbling

28 - +1 for calling this a great idea. The licensing hurdle has deprived the world of a lot of innovative and useful application ideas over the years (many of them mine). The notion of that hurdle going away has my brain spinning at the possibilities.

The first great app idea would have to be a full-featured user registration/management app to cope with the influx of new users to all the other apps that will now face outwards.

Secure customer self-service apps utilizing reader fields jump to mind next. Then you have premium publishing that includes a subscriber or "guest author" login. I'm currently exploring the migration of a Blogsphere blog I manage for a friend to WordPress or similar in part because of this kind of issue (he wants to greatly expand his author pool, but registering subscribers might have appeal as well). If it were not 2AM I could go on...

29 - I find it hard to evaluate this as a tactic without having a clear understanding of the strategy. I don't mean our strategy, but IBM's. If the strategy is to establish the Lotus brand as the dominant Enterprise 2.0 solution and to engage Micro$soft and others in a battle to defend/establish marketshare then it may make sense to use a product like Lotus Notes as a loss leader against SharePoint. In that scenario, eliminating any barriers based upon price to make the solution more appealing for extranet and external Web sites makes sense. We know Microsoft are doing all they can to eliminate prices as an issue in the short term. But IBM may have other plans. If I had a clearer understanding of how IBM plan to position Lotus Notes and Lotus XPages in the future I may have a better chance of evaluating the many great ideas (including this one) about how the product can be further improved. Executing our attack with laser precision may prove more effective and less damaging than pulling out the ol' shotgun.

30 - Some long comments here, no time to read them all, but absolutely add my YES, YES and YES again. The reasons for the yes are simple, obvious and will still be a yes if the IBM licence changes only apply to v852 and up!

31 - @22 - "...allowing Anonymous HTTP Access for Domino Express and it did not happen..."

I'm confused. Are you saying that anonymous use of a Domino-based website is not permitted without Utility server? I'm 99.999% sure that it *IS* permitted. A user without login credentials does not require a CAL, and the server can be public-facing as long as it's licensed like any other internal Domino server.

It's having the ability to authenticate that puts users into the realm of requiring a license. That license obligation can be met by either purchasing the "unlimited use" option of Utility server, or by purchasing individual CALs for every username/password pair in any directory. My objective here is to say "stop requiring a paid licenses when those username/password pairs are for non-employees."

The more I think about this, the more I'm skeptical about what the potential revenue sacrifice is for IBM. Putting aside sites that are internal to the Lotus community like ideajam.net, openntf.org and bleedyellow.com, how many Domino business sites can you think of that allow self-registration? Have you EVER seen a company that grants access to a customer by creating an entry in a secondary directory?

I'm thinking that I've seen this maybe 4 times in my almost 20-year career on this platform. And on the two occasions where I know Utility server was purchased, it was because the company's ENTIRE business was built around the application itself.

32 - Psst ... hey ... listen... a little secret... on Lotus Foundations you can have anonymous http access.
It is much, I mean a lot, cheaper then utility server.

33 - sorry, for Tim read Nathan

34 - @12 - I should be clear that I do not think the effect of the licensing change is that companies that currently have extranet solutions built on any competing technology would suddenly switch them to Domino. I think the effect would be that companies that already have Domino would start to use it to conduct business with trading partners. This isn't a net new license strategy. It's a depth strategy, making Domino more integral to how Lotus' existing customers do business.

In answer to your questions...

"Does NSF support transactions?"

Not yet, but this is closer than ever before with OSGi plugins that can bind to extension manager hooks. However transactional boundary requirements are greatly reduced with Xpages, as a lot of the traditional Notes "data synchronization" problems can be disposed of.

"Is the LS API equal to Java 100%?"

Yes, I believe this is the case now. I certainly haven't written any meaningful Lotusscript in about 18 months.

"Can I write field translation and field validation formulae entirely in a language other than @formula"

Yes. Java, Server-side Javascript or Expression Language.

"and be 100% searchable?"

I don't know what this means.

"Is recycle() dead?"

Tragically no. But the requirement for it is greatly reduced as Xpages automatically recycles all objects at the conclusion of every request. This has the unfortunate consequence of preventing Domino objects from being scope-cached from request to request, but this is only problematic in certain edge cases where you want to do thinks like maintain an arbitrarily built collection in a pager. It's a bit of a pain, but not nearly as much as it was in the 6.x era (where you had to commit the collection to disk as a Folder if you wanted to show it to a user.)

Do I think Xpages magically becomes the right fit for weather.com? No. I doubt there would even be a very compelling reason to refactor such a large, existing site to Domino any time in the next 5 years. But it rarely makes sense to refactor an existing successful application into a different platform, so I can't even imagine what any vendor could do to make you switch.

Unless it was to make the "Radar" option a lot more obvious when you put in a ZIP code. 6 months ago I couldn't find it at all, but at least now I can see that it's the next tab option after "month," which seems a little strange to me since it implies that looking at the current area radar is looking at a timeframe projection longer than 30 days.

But you don't need Domino to fix that.

35 - Add me as a yes on this, I'd be happy if there were just a small, medium, and large versions. The whole PVU thing is just a mess also, you start talking with a customer about that and they turn off even before you get started.

Glad Luke jumped in here, he just validates what the community has been talking about. Maybe if he knew about some of the enhancements that have been added to Domino, he might have taken another or earlier look.

36 - @8, Erik
Doesn't "The engineering there really is fantastic stuff, simultaneously simple enough for the newbie AND detailed-enough to allow an uber-coder..."

conflict with what you wrote as comment 39 of Ed Brill's post, { Link }
"But the bugs/problems/incompleteness of major standard Domino controls cause a significant barrier to legacy Domino devs trying to maintain their RAD-ness while learning a new technology."
?

37 - @11 - "Won't mean it's for EVERY app. Just the ones that are visible to the world."

How popular a platform is it for public-facing sites, ones that meet your criteria?

Cheers!

Luke

38 - @26 - is that the self contained self healing IBM product that needs a 3rd party add-in to do Domino web stuff ?

39 - @9 - Actually, coming back to it, I'm a little confused. The CAL-based licensing between Express and standard licensing isn't all that big a difference if memory serves. It's not like utility server, where the jump is an order of magnitude. So if it doesn't matter whether external users are accessing your application, then you're only talking about either 1000 or 1001 users, and I don't remember the difference being particularly large. I think the CEO license comes out to something like 20% more over Express, doesn't it?

That being said, my proposal would be that the license charge would be based on the number of employees accessing the application, not in the enterprise as a whole. If only 50 internal employees are in contact with customers, then you only need 50 licenses.

But let's be real, if you had 50,000 customers, and only had 50 people accessing the application, that isn't going to last. Other employees will want and need access to that application. It will grow inside your enterprise. Virally. Like Notes did in the 90s.

40 - @14 - "How popular a platform is it for public-facing sites, ones that meet your criteria?"

Right now it's not popular at all. That's the point. There are plenty of enterprises running Domino for enterprise collaboration, whether email or applications, who don't use it to connect to customers. Why not? It's ready-made for it. They already known how to deploy, maintain and administer it. There's hundreds of useful applications available for it. It seems like a total no-brainer to use it. It should be as natural for inter-company collaboration as Excel spreadsheets and email.

And yet it's not? Why not? One of the barriers is the license cost for an application that makes use of Domino's single biggest strength (integrated authentication & identity services.) It takes almost zero effort to remove that barrier. Certainly a lot less effort than would be required to, say, abstract the NSF storage facility to DB2. And the results turn out to be far more useful.

So it's an easy win for IBM and its customers. Is it the ONLY win? Hell no. But the reasons for NOT doing it are pretty much hot air unless Domino is doing a brisk business in utility server for extranet applications that none of us know about. (Note: they could be doing a brisk business in utility server for INTRANET applications in non-Notes mail shops. I really have no way of knowing this.)

41 - @26 - Psst, on any Domino server you can have anonymous HTTP access.

42 - @Nathan - Agree 100% about the licensing.

@Luke - Thanks for posting. I really, REALLY hope the appropriate people at IBM read your post.

There's still room for some more improvement on some of the items you mention, but fortunately the "catch up" put in with XPages over the past couple of years addresses many of your pain points. The engineering there really is fantastic stuff, simultaneously simple enough for the newbie AND detailed-enough to allow an uber-coder to "dig deep" for extreme levels of granularity and control.

43 - I've been pretty vocal about this for a while. PVU licensing is evil { Link } and Utility server (read Extranet for anyone familiar with IBM but not Domino licensing) is a mislead approach at best.

However, let's not discount how much revenue IBM makes from utility servers. I'd guess the bean counters would ask (and this is what I would do in my business) , "If I remove the *extranet* restriction where will this revenue and margin be replaced from?". You could argue the "free it and they will come approach" will work (and if you think this works you need to read { Link } ) and hence this will drive more revenue. This *could* happen, but doubtful. I think it may drive *visibility* but this could be a good thing or a bad thing (I think any publicity is good publicity, hence my "Darren in a banana-hammock picture series on flickr), but IBM's outlook my be different. Still I guess if you "free" it at least the current customers would be happy.

Going forward it is going to be interesting to see "what" IBM do, or more importantly "if" IBM do anything to change the way this is going. Either way, the pricing of Utility server needs to change. The outcome (or lack thereof) is going to tell you an awful lot about IBM's intentions with Domino as an application development platform.

44 - We could have done several extra Domino deployments in years past were it not for these licensing terms. So agree 100%!!!

45 - I'm rushing off so just a quick comment for now.

I think that Tim's idea is a very good one.

My only caution is that I was thinking about deploying Xpage Apps non domino ( i.e. "net new" ) customers. If my "net new" domino/xpage customer has 30 ish people using a 10K app that makes it more expensive than using the cheapest version of utility express - although it would be nicer in the respect of not needing to worry about PVUs and finding 1 cpu servers to run it on - but lets not go there for now.

So yes this makes a huge amount of sense, is easy to do and surely if anyone is supportive it will be those existing customers who would really welcome some good news re licensing. Just lets be careful not to hit the "net new" Xpage applications market which interestingly would get a great boost from more Xpage apps out there generally.

As I said in the original article I don't know the answer. What is obvious is that IBM needs to make a good return on whatever scheme it chooses. My wish would be that with "net new" Xpages they recognise that a valid approach is lower licencing cost but more licences over all, otherwise this great opportunity that is Xpages will be crippled from the start.

I suppose that the problem that I am facing with "net new"xpages is that as the provider of the first "Xpage" application my projects total cost must include the infrastructure.

As I said, not sure of the answer but support this particular idea and the debate in general.

Anyhow, must run, Sean

46 - @10 - I'm talking about the general RADness of XPages as a way to write compelling interfaces overall (which it does extremely well) while allowing you to delve all the way down to the complexities of the JSF implementation. But only if you want to. The problem I was describing on Ed's blog is that to recreate the native legacy Domino widgets in XPages you need to "roll your own" in some cases.

Which is absolutely 100% doable, and it's not incredibly complicated, but it's not as simple as dragging-and-dropping an embedded view control onto the XPage, like it *should* be to help convert old-school Domino devs.

47 - Luke, might I suggest you download Designer 8.5 and take it out for a spin? It's free now. I suspect you'll be impressed by what XPages are capable of.

48 - @9 I'm not saying it solves every problem. Just the one about whether domino is for apps. Won't mean it's for EVERY app. Just the ones that are visible to the world.

49 - Well for years I am begging for at least allowing Anonymous HTTP Access for Domino Express and it did not happen. You still need Utility Express or an Enterprise server to achieve this.
I really like Luke's comments because it very well fits to what I see nearly every day. The world is not waiting for Domino and to get back some of the attention I fear even free Extranet licenses won't be enough.
I also think IBM has already removed some barriers in this area. They just prefer it to happen on Lotus Live. At least I often read that communicating with partners and customers is much easier (cheaper) there.
Of course this does not mean that IBM should not at least give it a try. It is a great idea but I am curious what the Websphere and Information Management divisions would say about it. If the price is that Domino is not allowed to innovate (like when IBM realised that Garnet was getting much to powerful for this platform) then I vote against it.
I currently have two projects that deal with moving from Domino to another platform. I am sure at least one would be put on hold immediately.

50 - @22 & @23 here is the license verbiage:

"An Authorized User is a unit of measure by which the Program can be licensed. An Authorized User is a unique individual inside or outside of Licensee's Enterprise with a specific identity that is validated when accessing a IBM Lotus Domino server. The Authorized User's unique identity could be defined by his or her Notes ID/password ("PW") combination, his or her IBM Lotus Domino HTTP Name/PW combination, or any other third party authentication source that defines the Authorized User's unique identity."

From 8.5.1 license agreement { Link }

I didn't stay at a Holiday Inn Express last night, but that would seem to validate Nathan's theory about authenticated users forcing the need for a license.

IBM? Care to elaborate?