Campaign to Open Source IBM's Notes/Domino 255
Ian Tree, an IT consultant from the Netherlands, has started a campaign to convince IBM to open source the code for Notes/Domino. Hoping for results similar to the push for Sun to open source Solaris, which finally saw success in 2005, Tree makes the simple point that it won't happen until someone asks. "By being an open source product, Tree is also hoping that Domino becomes something schools use to teach groupware and application development concepts, which is the holy grail for future market adoption. This is how various Unixes, relational databases, Linux, and a raft of other products eventually became commercialized. While the idea of open sourcing any proprietary program is appealing, in as much as it sets a program free to live beyond the commitment (or lack thereof) of its originator, it is hard to see why open Notes/Domino would have any more impact than OpenSolaris."
CouchDB (Score:5, Interesting)
CouchDB [apache.org], which has been generating some [eflorenzano.com] hype [blogspot.com] lately (especially among Rails fans), is by Damien Katz [damienkatz.net], who did work on LotusNotes and Domino, and claims CouchDB is inspired by that.
According to him, Lotus got a lot of things wrong, but it got the database right.
I don't know if there would be anything to gain from the original (even just to read through it), or if we should all be focused on CouchDB now, but it would be interesting to find out.
Companies open source unprofitable products (Score:3, Interesting)
As I state in the title, companies only open-source unprofitable products. As I understand it, Sun was willing to open-source Solaris because it was no longer profitable by itself - instead, it was just driving sales of Sun hardware. Until I see some similar evidence regarding Notes (showing that its unprofitable on its own and only drives sales of other IBM products), call me a skeptic of this effort.
Re:As an Outlook/Exchange fanboy.. (Score:4, Interesting)
I fail to see how making Notes open-source would help this aim. After all, the main obstacle to people using calendaring and groupware apps is that said apps are difficult to use. Given Notes' horrible record regarding usability, I fail to see how making Notes freely available to all would spread the usage of calendaring amongst the general computing public.
In fact, I think that GMail and Google Calendar are doing more to spread automated calendaring than open-sourcing Notes (or even Outlook, for that matter) ever could.
Re:What we want? Isn't it? (Score:2, Interesting)
IBM will never open source it (Score:1, Interesting)
Unlike Sun, IBM doesn't believe in OSS one tiny bit.
IBM believes in a future monopoly and the money could bring.
The only things that IBM has made open source are complete crap, like SWT, Xerces, Axis and the likes.
Hell, Eclipse wasn't even half useful until years after it became open source.
IBM has only open sourced things to kill competition, ever. As Notes Domino doesn't have competition, if you look form it as more than an E-mail system, it will never happen.
Fuck IBM and their hidden agendas. Fuck WebSphere, DB/2, ZSeries and all the rest of their crap.
I can't really say anything bad about iSeries though, which bugs me, as it's really cool stuff. Maybe because they're made in Norway, of all damn places.
Re:Slow news day? (Score:4, Interesting)
The Notes/Domino product line generates somewhere in the neighborhood of a billion dollars a year for IBM in pure software sales (not services.) It's also recorded 15 consecutive quarters of double-digit growth, and has grown by over 50% since 2004.
You can see more at the long-running blog of Ed Brill, former worldwide head of sales for Notes/Domino, and currently Director of End-User Messaging and Collaboration. He just finished a year-in-review post http://edbrill.com/ebrill/edbrill.nsf/dx/2008-the-blogging-year-in-review [edbrill.com]
Re:Why such hatred? (Score:2, Interesting)
Re:What was it before there was Firefox? (Score:1, Interesting)
It is not like IBM losing huge amounts of money or lost the market to MS Exchange because of Lotus Notes. If you remember the exact time which AOL decided to open Netscape to public, it is not like Netscape was really popular anymore.
IBM makes billions thanks to Lotus Notes clients and server agreements. Their clients seems to like Lotus Notes or Notes wouldn't exist today. What we got here is, dozen of people popping up on Slashdot, the very same people each time bitching about it. If you ask them, Exchange sux too. Also they claim postfix isn't scalable while Yahoo etc. with 250 million users happily use it.
AOL opened the source because they had no clue what to do with it. Perhaps IBM customers will all run to MS Exchange hands if Lotus becomes the horrible buggy kitchen sink written by amateurs. Remember pre 1.x Mozilla?
Re:The database is the problem with Notes... (Score:3, Interesting)
I seem to remember best describing it as a slightly multiuser filemaker pro flat format with a lot of hype to rip off IBM for a few billion dollars.
I'm really not sure what it's got to do with FileMaker Pro, as I don't know a lot about Notes. I could compare FileMaker with CouchDB, though:
CouchDB is schema-free. There are no predefined columns. Each record is any JSON object, with no constraints other than that it be valid JSON.
Older versions of FileMaker are, indeed, one table per "database" -- but it was very much a fixed-schema table. Each record had exactly the same fields. I've seen more than one nightmare database which would have been vastly improved by the concept of relationships and multiple tables. Granted, you could relate two "databases", but this was fragile and not for newbies.
What's more, FileMaker requires that you be connected to a central server, and somehow manages to be dog-slow over a network. By contrast, CouchDB can do asynchronous multi-master replication -- to the point of completely disconnected operation. As in, you could run a Couch server on your laptop, take it on the road, connect later, and it would deal with conflicts appropriately.
if it had had a good database to begin with, then, 3rd parties could have salvaged a good groupware database product with add on tools or even clients.
I don't know -- how accessible was the database? Given that it wasn't free, and was tied to Notes (which has a horrible reputation), it's not surprising that people didn't pick up on the database.
Keep in mind, good ideas aren't necessarily adopted right away. There are a lot of strange things about Erlang, but it does scale very well, both to cores and multiple machines -- yet it hasn't had a lot of adoption compared to, say, Java. What's more, there's a grand total of one other language running on the Erlang VM, and as far as I know, no other languages built around the same concept -- if it's really the syntax and weirdness of the language, why not steal the best parts of it?
I could pick other examples, but I think I've made my point: A lack of popularity doesn't necessarily mean something is bad, any more than a lot of hype necessarily means something is good.
Ugh (Score:3, Interesting)
I wouldn't use it if it was free!
I had to use that POS back in my days working for a DoD contractor.
Who cares if the underlying db is sound, the client exposed way too much of the db and as a result was a user interface clusternightfuckmare. Yes, it was so bad I had to make up a word for it.
Even Outlook with it's Russian Nesting Doll configuration options*, is a way ahead of Lotus.
*If you're not sure what I mean, follow these instructions for an extreme example from Outlook 2003:
Tools->Options->Mail Setup->Send/Receive...->Edit->Account Properties->Advanced->Remote Mail->Retrieve items->Filter->Advanced.
You'll now be six modal dialogs deep in it's options, past two Advanced buttons!
Further, did you know it's possible to change your domain password from within Outlook's nesting dolls? I'm not kidding! But good luck finding it.
Re:CouchDB (Score:3, Interesting)
Having lambdas and eval is not enough to be a Lisp dialect
No, but it's enough to be quite a bit ahead of many other languages. For that matter...
Lisp is about S-exprs and code-as-data
Granted, it's not as clean, but most Javascript implementations include a toSource method on the function object. Is that enough to build what you want?
S-expressions.
Somehow, I like the JSON syntax better. But it's also a reasonable representation of a data structure, such that I would think it would be easy to store sexps that way.
side-effect free, lenient evaluation, no guaranteed evaluation order - all very important to enable advanced query optimizaters.
In other words: Performance. Got it.
You realize that CouchDB scales to an arbitrary number of nodes, right? And while the queries don't yet, they easily could, and will in the future?
Yes, when I know an app is going to be on a single machine only, I'll probably use something like sqlite for the raw speed and ease of deployment. But I'd much rather have something horizontally scalable than vertically scalable -- and as a nice side effect, I get a decent imperative language that's close to what I already know how to use. And if I really have a problem with JS, I can use whatever language I want.
I could, for instance, serialize sexps to JSON, and write an engine that deserializes them and passes them to LISP view functions.
Contrast this to SQL -- maybe faster on a single machine, but how do you make it really scale? How do you know what to index, when? How do you scale your queries to thousands of nodes?
There's a certain class of application for which SQL makes perfect sense. There's also a class of application for which SQL is a constant pain in the ass, and CouchDB makes perfect sense.
having first-class functions doesn't change that. Yes, you can go halfway there by using lambdas and such, but JS lambda notation is really way too verbose for that sort of thing
So write some sort of macro engine. In fact, someone is doing that -- check out jabl [github.com], from the author of haml [hamptoncatlin.com].
Yes, it's ugly that you have to do that -- but then, I find LISP syntax ugly; unnecessarily verbose, and homogeneous enough that it's hard to tell, at a glance, what's going on.
I don't think Couch is ready -- the two features I really want to see are a nice Ruby interface for views, and support for stacking views. That said, I doubt I'll be using SQL for much even a year from now.
Comment removed (Score:3, Interesting)
Re:Open source and Lotus Notes? (Score:5, Interesting)
Disclaimer: During a different stage of my IT career, I was a certified Lotus Domino Application Developer -and- System Administrator.
The Lotus Notes UI WAS overdue for a significant overhaul. For years, it wasn't horrible interface design, but LACK of design that led to the meandering mess that most people experienced in the last two decades.
As of August 2007, IBM finally released a truly well-designed Lotus Notes mail client: Lotus Notes version 8.0, which is, IMHO, the most comprehensive remaking of the Lotus Notes client and its e-mail interface since Notes began. Every client release up until now had UI changes that were evolutionary at best.
The new client itself now sits on top of the Eclipse Rich Client Framework, and will consequently run on Windows and Linux (Mac support coming shortly with 8.5). And you can still access all the same Lotus Notes corporate applications that range considerably in quality. And in fact, the Notes 8.x client can still access Domino 7.x mail files, and they will look exactly the same as they did before (although client menus have changed).
But if you run Domino 8.x servers, with the 8.x mail template, and are using Notes 8.x, the e-mail UI is a ground-up redesign that is far superior to anything that came before it. If you've ever whined about Lotus Notes mail in the past, you should check it out - that complaint is now outdated.
My 2 bits...
Re:The database is the problem with Notes... (Score:3, Interesting)
The Notes "database" was one of the strengths and weaknesses of Lotus Notes.
Background - speaking strictly about the native Notes Storage Facility (.nsf) format (and not to newer options for RDMS virtualization or or DB2 backend for custom development):
- Everything in Lotus Notes is stored in a "note", each of which has an XML-like data structure. Keep in mind that there was no such thing as XML, or even the internet, when Lotus Notes was first developed.
- User data notes are usually called "documents".
- The components of the design of Lotus Notes application are also stored as XML-like data notes, and are usually called "design elements". So even forms, views, images, script libraries, etc, are stored in "notes".
- The data notes are not dependent on the underlying design elements for their existence. This is a really weird concept for SQL DBAs, where the data is bound to tables for its structure. Drop a table in the SQL world, and you lose your data. In the Notes world, if you delete a field from a form (a design element), the form itself, or even delete views (also design elements)... and the data is unaffected.
- All of the data and design "notes" for a Lotus Notes "database" are stored together a self-contained NSF file.
The loosely structured nature of a native Lotus Notes database means it is both VERY flexible, but lacks some of the rigor (and related benefits) associated with a true RDBMS. Oh, and you can't use SQL to do cool things like left and right outer joins. Instead, you use a Lotus formula language, LotusScript (very similar to Visual Basic), or Java to "lookup" data for display or for repeated storage within a note.
Even Notes data types are much more flexible/loosey-goosey than found in the RDBMS world. They can be boiled down to:
- Text (stored as the equivalent of varchar(32768) in the SQL world)
- Numbers (no need to define integers, floats, or doubles)
- Date/Time
- Rich Text (including attachments, formatting, tables, etc)
- "Name" type fields, which are related to Lotus Notes security.
The loose structure also lowers the barriers of entry to slap together a Notes database. A person can know enough to be productive/dangerous without having a clue about referential integrity, primary keys, or tables.
No, the database does suck. (Score:3, Interesting)
My problem with Notes was that a lack of a relational structure made it awkward to do something like a document management system in it, where you would want to have a table of authors recipients and other persons associated with documents in a relational sense. The hope was that you could use a notes database to represent the rich document stuff, which it could kinda do, but also, have some sort of a relational, at least more strongly typed nested collection representation with it and you simply couldn't.
You would want a nested list of authors to be well, authors, of a first name and a last name and other useful information on them, and the same with recipients, cc's, and so on. From there you could build a timeline of who saw what and wrote what correspondence about what topic and that would give you the facts of the case as a simple select. But you couldn't do that, and, at the time, Notes even had problems with just nesting tuples, period. So yeah, Notes database sucked.
And the Notes idea wasn't even novel. The idea of a Notes database being a packed record, indeed, the whole JSON concept of rich data stuffed into a database, was tried before Notes, before XML, way long ago using a system IBM developed called "PICK". PICK was an interesting hybrid in that you could stuff rich data into a field but you could also use a forerunner of a SELECT command, called LIST, to fish stuff out of it.
Re:uh, no? (Score:3, Interesting)
Has it really failed? Don't companies still pay IBM lots of money to use it?
Yes, they do. My company (Humana) is one of them. If it were just a matter of switching email systems, we would have gone to Exchange long ago, but we have hundreds of workflow forms and other crap built on Notes databases which are unfortunately integral to a lot of business groups.
Never in a million years could I have imagined I would be longing for Outlook and Exchange until I started at this company!
Open Source This... (Score:5, Interesting)
Some of us would like to see IBM "open source" other stuff, like OS, VSE, VM. Heck, they can keep the source, just give us a hobby, or a "not for profit" license...
And they can keep the current stuff, the stuff they make money on today - like z and ESA. I'ld be more than happy if I could run something like the 20+ year old VSE/SP on my PC at home, under Hercules...
But no, we're stuck with 40 year old "public domain" software, stuff like DOS 26.2 from the System/360 days. Hey, it was fun, I was the "Sysgen Kid" back then. But it only remotely relates to what an, even 20 year old mainframe, is all about today, stuff like CICS...
CICS is a perfect example. Back in the 1.x days, you had 100% of the CICS source code, as long as you had a license. Today "transactiuon server" is a big secret...
Sorry for the OT rant, but it ticks me off. IBM has dumped billions into Linux, but us old greybeards, those of us that wrote those countless lines of Assembly and COBOL and RPG, and yes, CICS code, custom code, without which IBM would have a great OS and nothing else. Those of us that worked shift after shift of unpaid OT, tweaking that demo, making it perfect for the guy that will be spending the IT budget. Those of us that helped make IBM what it is. Those of us that truly enjoy what we do, as a job and hobby...
We can't play with our toys at home, legally that is...
I'm going to retire in a few years. I won't be a licensed user any longer. And I surely can't afford the 4 figure monthly "commercial" software license fee, let alone the 6 figures to "buy" it...
Come on IBM, great "open source" promoter that you have become lately. Do it for us original geeks, we need something to do in our old age...
Open source this - VSE/SP 3.1