Damien Katz

Or alternatively we can explore WHY management doesn't like Notes and work with them to demonstrate the very real value in the product that they already have. It's sure to be more productive and cost-effective to use the existing Domino infrastructure than to move to a less mature, less capable product down an undocumented migration path.

Couch DB may be a good product in its own niche, but apart from your claim that some management doesn't like Notes I see no advantage in recommending your platform to my Notes customers - especially when the customer will wind up supporting two architectures and still needing to pay for their Notes licenses.

Let me get this right: they like CouchDB because it's so similar to NSF but it's not Notes?

I've got to agree with Graham, this time/effort would likely be best spent discussing the pain points with Notes (like, are they on 8.5.x?) And how will they handle mail? C&S? Directory management?

Either way, wouldn't your employer (IBM, right?) be a great place to start? They could get either get engaged on the education/customer-retention front, or get involved at some level working with (or even creating from scratch) a "Migrate to CouchDb" team.

I've been following your CouchDB work since day one, and I'm a fan. But there is simply so much that NSF offers over it that I'd be shocked any shop using it would even consider swapping for CouchDb. I could see choosing CouchDB for a project if you *don't* have Notes/Domino, but swapping just seems... ignorant.

LOL. @All Damien isn't an IBM employee anymore.

How about migrating people from Oracle MySQL, first?

Without causing a flame war I have seen this as well. Senior management want iPhones and they want Outlook. Forget all the database functionality.

The other thing they are looking at doing is divesting the cost of managing 1000's of email accounts - GMail, Hosted Microsoft Exchange etc.

The email component can always be migrated to Outlook but like Damien is suggested all those legacy Notes databases need a home.

Damien already said it. There are companies that want to move away from Notes so why not move to CouchDb instead of another platform?
Selling Notes and Domino is not the job of Couchio.
Notes is broken for many. You probably can't fix this with plain talk otherwise companies like Quantas would not be moving.
And if you give IBM money they will move you to any platform you like, even Exchange and Sharepoint.

Nobody wants a flame war and I don't see one starting here.

My question is what are the advantages of moving from Notes to Couch DB. So in Damien's opinion some management don't like Notes but that doesn't justify moving to Couch. Forget about Qantas and IBM... just tell me any technical advantages Couch has over Notes. So far I haven't heard any.

Graham. CouchDb is not Lotus Notes and Domino. It is just a part of it. You would need to evaluate the full package and those can look very different. You can use CouchDb with a lot of platforms.
If you compare CouchDb to nsf you will find similarities. If it works like nsf but is free of Extranet licensing, Processor Value Unit Nonsense and an open platform that alone makes a difference.
It also does not have a 16Bit database core so you probably do not have 16kB or 64kB limits. It's core is much smaller and you can install it on a lot of devices. In theory it could probably run on mobile platforms like Android too (I have not checked this). You won't have to rely on IBM to support your platform.
It does not have a 64 GB limit (at least not that I am aware of). You can use DAOS on Domino but this does not work for local replicas.
You can probably find a lot of points where Domino shines too. 20 years of history is an asset and IBM would have done something wrong if a small company could repeat this in just a fraction of time. So should you move from Domino to CouchDb?
Probably not but if you already decided to move then you may look at CouchDb. Couchio has just started.

Henning, thanks for the update. I think that any customer who decides to 'move off Notes' but doesn't have a firm destination already in mind is making a big mistake ... and I will politely tell them that.

If product XYZ gives a better business and IT solution for your company and you can swallow the migration dislocation costs then go for it, but if you are moving 'anywhere but Notes' just because you hate Notes then you are not fulfilling your obligation to your employer. I have yet to see a job description that includes the right to migrate the company to a technically less capable product simply because it makes you feel good.

Well IBM cancelled Domino.doc because nsf seems not a good place to store documents.
They put Domino Workflow in maintenance mode years ago.
QuickR is moving forward very slowly (with the 8.X codestream starting in 2007 and having 8.5 promised for 2010).
They made the Sametime Gateway run on Websphere/DB2 because they said that Domino is not powerful enough to handle it efficiently.
Now Sametime is put on hold on the Domino platform too and the product itself is moving forward in IBM's J2EE offerings.
They invented Lotus Forms because Domino even does not seem to be an efficient handler for form management.
There is Lotus Web Content Management because Domino can't do that either.
They bought Outblaze for general purpose PIM offering with more than 18 million accounts in Lotus Live because, guess it, Domino can't handle it (to be fair Exchange can't handle hotmail either which is using MS SQL as the data store).
IBM built Connections on J2EE Websphere / SQL. And many third party vendors are giving up on Domino too. It now happens that you have to wait months to get support for certain AV products or backup solutions. IBM answers this with their own offerings instead of working with those vendors to get releases out in time. When I talked with Kaspersky in December last year they said I was the first to ask for Domino support that year (Kaspersky sucks anyway). Lotus Protector for Mail Security is an appliance that does not use any of IBM's Domino technology.
Now when IBM does it, it is a well thought out strategy, if their customers do it, it is clueless.

We are software development company and we have Lotus Notes development team. And in recent years we have a problem that it is quite hard to sell Lotus Notes/Domino based solutions to new customers (as they already have some e-mail / calendar solution) as they are not eager to have one more proprietary platform. And as we are also thinking of providing more solutions as software as a service then we also are interested in open-source platform to reduce our costs.

Therefore we started to learn CouchDB and plan to do some initial pilot applications. CouchDB document model is very similar to Lotus Notes document model therefore it is quite easy for our Lotus Notes developers to understand it and to port data from our existing databases to it. Of course much harder part is UI development - now they need to learn much more JavaScript instead of LotusScript. We are going to do our pilots using CouchApp and related plugins.

We will let you know when we will have some progress :)

My apologies Damien. IIRC somewhere on your blog yesterday it said you still worked for IBM full-time on CouchDb. I hope it was an amicable departure.

Either way, congratulations on your new company!

Raimonds If you work with Domino you would have to learn Javascript anyway as Lotusscript is not moving anywhere and IBM's new (well it is not new at all) platform XPages is all about Javascript (and Java).

CouchDB must be

CouchDB must be the most suitable middleware for migrating from Lotus Notes/Domino system.

The data migration is not so difficult except RichTextField in Notes.

But Notes/Domino systems often have several business workflows integrated with other systems (file/mail transfer, approbal signature, ...). I think it would be more suitable if CouchApp (or other tool around CouchDB) had business workflow framework.

I do not know which BPEL or JavaScript is better for the framework....

OK, I'm a admin, and I think this is a cool idea. Just make the tools easy enough for me to snap in on-site and run, convert, then maybe have a programmer tweak the workflows afterwards. I tend to think that we over analyze how much functionality we need to bring forward from the Domino apps. Most people want to get off the Domino platform, they want to get the data and move on.

We can debate the value of Domino, but I relate it like a car you know your going to trade off on something new. You can have the mechanic tell you that it will only be this much to fix it up, but you won't be happy till the new wheels are in the drive way. Businesses just don't want to invest in the Domino platform anymore for applications, and I for one, am tired of trying to talk them into it, time to move on.

If you want an Admin/implementor opinion of the tools, give me a shout, I'd be happy to take a look.

Good Luck and Happy coding, look forward to seeing the results.

Bill

The part that I find intriguing is the mantra from MS Exchange (and now CouchDB) consultants which basically says that their product can't do what Notes/Domino does but you should use it anyway. Sure, Domino has technical limitations and IBM needs to fix those problems, but unless your replacement product completely replaces Domino (which you admit it can't) then why on earth would people use it?

Can someone explain the cost-justification for running both CouchDB and Notes/Domino... or are customers supposed to just throw away their existing Notes LOB applications and go back to paper and pencil for recording data?

I'm not annoyed or flaming anyone here, but surely you agree that there's no financial benefit or technological sanity in running multiple products(CouchDB and Notes) when one product(Notes) can do the job. Henning listed some of the issues with IBM's support for the Domino product line, but I really can't see CouchDB (who are pleading for FREE assistance from other developers to write their documentation and work out the migration issues) as a better strategic option.

Good luck with your product guys, but I think I'll wait for a few more versions to hit the streets before I look at it in detail.

Exchange consultants don't say this. For the PIM part Exchange is competitive so consultants would probably say that they are much more advanced. Microsoft has another strategy here. Microsoft could also ask why there is no Finance proven ERP software for Domino as Microsoft has Navision/Dynamics.
CouchDb consultant probably don't say that either (I am not one). They say if you want to move off Notes but like the datastore (that just makes one part of Domino) you could take a look at CouchDb. There isn't even a CouchDb mail server yet although I consider both nsf and CouchDb an ideal candidate for storing emails.

What is the cost of running CouchDb?
Well the license itself is free, dedicated support is not. You also do not need dedicated servers Couch will run on any existing server. You could even run it on the same server than Domino. The majority of Domino customers do not use it for their public website. So everything on Notes and Domino is not a reality anymore. Not even mentioning what you will add if you run Domino but want to use Connections or Sametime Advanced or the IM gateway. Extra hardware, extra servers, in many cases completely different software (J2EE, SQL) but of course this does not matter because it is IBM software. You cannot even run Domino Sametime and QuickR on the same machine (without using VMs).
And FREE assistance is what Open Source is about.
IBM does that too. They want you to write their Redbooks for free. They run Wikis that people should use to write documentation. They run Notes.net where users help each other with limited support from IBM people that spend some of their free time there. The difference is that IBM does more than 100 billion $ of revenues while Couchio just gained 2 Mio. Venture Capital. Competition is a good thing and IBM not a charity organization. They have all the money and resources to make Domino shine. So far people still beg for icons that have more than 16 colors (and 8.5.2 finally has them).

I don't have a problem with CouchDB as a concept - go to it guys - but IMHO migrating mission critical Notes apps to this environment is asking for trouble. There are minimum requirements that define an application as ready for Prime Time in the corporate world. Linux is getting there/already there (depending on who you talk with) but as far as I can see Couch DB isn't even off the starting blocks yet. CouchDB could prove valuable in three years or so, but IMHO moving live Notes apps to that platform is like migrating your company to Linux back in the 1990's - more emotion than sense.

It seems you are defining Notes as a replicating datastore just as Exchange consultants see Notes as just email. They offer to migrate mail users to their platform and they just ignore the other Notes services. CouchDB seems the same kind of game - we'll migrate your replicating data store to CouchDB and ignore your:

* need for email - just keep Notes running :)

* need for programming - you might despise Lotusscript in Notes but it's there and it works and so does Java and Javascript.

* need for IT stability - you really want me to run this beta software on an existing production server? I don't think so! If I'm going to implement it for a customer (which I'm not) then I'd recommend a sterile server which can't affect other systems.

* need to work with existing Notes data - maybe you can migrate the Notes data, but those imported documents are pretty useless until you build a CouchDB app to service them.

* need for product maturity - IIRC IBM pumped a billion dollars into the R8.x release and you're working with one five hundredth of that budget. Sure IBM has screwed up in trying to implement NSF2DB2 and other flights of fancy, but after twenty years of development and backward compatibility they are technically light years ahead of the CouchDB team and I don't see how you can narrow that gap.

* need for worldwide support - what is it? three guys and a web-site plus some dedicated consultants? I have no doubt you guys are all far smarter than me, but if you are doing all of the support then who's writing the code for the next version?

* need for integration with Instant Messaging - yes, there are Sametime server issues but it all integrates at the front end for the user.

Against all of this you're complaining about 16-bit color for your Notes icon??? Your complaint is valid but fairly trivial.

I agree that IBM is not a charity and I've been ripped off by their pricing policies more than once, but at least they are not asking consultants to write basic product documentation ("How to migrate to our platform") for free. I see Redbooks as an optional luxury item and I love it when a new Notes one is available.

Summary = I love the idea that Notes has potential competition but beware the fate of the frog who tried to blow himself up to the same size as the Bull.

Fanning the flames :O

http://google.com/trends?q=couchdb%2C+lotus+domino&date=2ytd

Great -- migrate to CouchDB, make some money providing services. Migrate back to Domino in a couple years, or the next collaborative solution -- make some more money providing services. What business value do any of these migrations provide ? None -- it just provides income to service providers. It's not the software, it's the business processes that need improving. Notes/CouchDB/Sharepoint -- doesn't matter.

@Robert - These migrations provide a lot of value to the migration consultants. ;-)

@Damien - LOL!

Searching for "Lotus Notes" is a bit different, though playing with the time range shows some very interesting and apparent changes in the derivative since the Notes 8 release. Based on the graph of the past 18 months I dare say 2010/2011 could actually show an increasing rate with the Lotus Knows campaign - unless IBM screws it up.

Seriously, I'm a fan of CouchDb. I think that from a theoretical standpoint it's got some great applications in a few environments. I hope that IBM uses it for ideas for NSF, and I hope it puts pressure on them to continue to improve that area of Notes/Domino.

But just going here:

http://wiki.apache.org/couchdb/Breaking_changes

...*should* be enough for any business who is using Notes/Domino and considering a migration to CouchDB to say "Hmph. Better give it a few years." If not, well, I guess they'll learn the hard way.

I am not sure if everyone in this discussion is aware that Damien rewrote the Lotus Notes Formula language in R6 (and most say he did an awesome job). I think he knows what Notes and Domino can do and what not.

@Graham You say CouchDb is immature but you have not even tried it? That is what I call a fair review then. Open Source does not hide their problems. If you want some magic then look at a few nsd files. Closed Source just hides them but this does not mean they are not there.

About a trivia with 16 color icons (and I hope it did not take a billion dollar to get it). I am not sure if I should write my list what is wrong with Lotus Notes. I think I would hit the 16kB text size limit. IdeaJam.net is not a bad place and I think there are hundreds of ideas to make Domino and Notes better.

CouchDb is not for everyone. It is not even a replacement for Lotus Notes. But a lot of companies leave Notes and Domino. If you did not like Damien's graph maybe this one is better http://dominoorexchange.pbworks.com . Even big companies have retired Notes and Domino as their premier collaboration platform. And guess what most of them probably do not even use CouchDb.

I apologize for having started a flameware. I promise that I will stop now (here) but it was great fun and at the end it is just software so no people were harmed during this discussion (hopefully).

Henning, apart from Damien and Erik throwing URLs at each other I see no flaming in this conversation so don't feel worried.

I am not claiming to be reviewing CouchDB so I don't see why you consider my comments unfair - my point is that moving to Couch seems to come at the cost of many of the features that I count as essential in Notes so I have no desire to go down that track. Surely there should be some feature in this new product that is better than Notes but I can't see it. Tell me what I'm missing - what can CouchDB do that Notes can't (leaving aside trivialities like 16bit color and size of text fields). CouchDB may well overcome some of your own pet peeves about Notes but its a certainty that CouchDB will generate its own peeve points so soon all the CouchDB consultants will be complaining to Damien "You can do xyz in Notes but not in CouchDB - when you gunna fix it?"

I have no problem agreeing that Damien and his merry men are a few orders of magnitude smarter than I am but just because they can write a replacement for *one* part of Notes architecture it doesn't mean that the world needs what they are building. Like I said, more power to their efforts because competition for Notes will spur IBM onto better things, but as far as I can see from the URLs provided CouchDB is as mature and capable as the beta version of Release 1.0 of Notes with a much smaller organization behind it.

Whether or not the top 500 keep using Notes or migrate to the MS stack is a different discussion - call me back when one of them migrates their mission critical apps from Notes to CouchDB.

Hello,
I like the idea. Unfortunately, I'm not working anymore as a notes developer but I will follow your story.

regards Rene

I think that before we have a setup file or a prebuilt setup package in a .zip file. You will only have a limited spread and if you want to attract Domino developers this is a must.

Duh? The Windows Installer for 0.10 works like a charm for me. The typical Domino Developer neither use OSX nor Linux (although Debian has 0.8 in stable and 0.10 in testing and Ubuntu for sure has it too) because there just is no Domino Designer for these platforms.
Get it here
http://wiki.apache.org/couchdb/Windows_binary_installer
0.11 hasn't been released yet although I think you can check it out anytime.

@Bill Dorge I am primarily and admin also and my development has primarily been for departmental type Dbs using mostly formula language (I started with Notes at v3) and a little bit of re-used code from the internet. What is a non-starter for a guy like me, sad to say, is the lack of any high level UI for dev tools or admin tools.

From my understanding Couch needs a business layer to solve common problems like: email templates, sceduled tasks, call to web services, interoperate with othr systems, resize images. I think the future is couch and a layer with dev tools will arrive. Just a little soon now.

When people wrap their cars around poles many of them blame the car, or the tree that jumped out into the middle of the street.

Stupid to compare the safeness of cars without looking at who's driving them.

As some have said here, its not the platform, it's the morons driving it. DOS batch files can be made to do some pretty fancy sh#t!:)

What I want to see is CouchDB replacing the NSF layer.

From what little I've read about CouchDB, security (lack thereof) seems to be something of a stumbling block if moving from Domino.

There's no document level security is there? No Reader, Author or Editor level access to databases either? And without those things, you don't have any kind of replacement for Domino, IMHO.

I'm happy to be proved wrong though because I do like the concept behind CouchDB.

Hello, I've been Notes/Domino developer for 7 year and now I'm studying CouchDB as an alternative NoSQL database and I really, really like it. And it was easy enough for me to wrote such tool so every one could get it here for free:
http://www.smartlotus.com/smart-notes-2-couch.html
Enjoy it and leave some comment while relaxing :)