Lotus Notes was the future of communications, a decade before laptops had WiFi. Yet of all things, it wasn’t even an email app. It was a notes app, a collaboration tool, an all-things-to-all-people software that let you build apps in the way Access and Airtable later would. That, and the notes could be used for email.
Love it or hate it (and there were plenty on both sides of the fence), what you couldn’t do was ignore it. This email-and-everything-else platform showed what the future of digital communications would become — and provoked, as email itself was always doomed to provoke, equal measures of awe and exasperation.
↫ Matthew Guay
I have no experience with Lotus Notes, but I do have some vague memories of the software being used at my parents’ employers back in the late ’90s. Note that Notes still exists and is in active development as HCL Domino (server) and Notes (the client). If you really want to, you can still run your company of office on Notes.
I wonder how many actually still do.
I loved Notes, I was gutted when the company I was with moved on to Outlook.
But there again Outlook, or any email client was better and certainly easier for 99% of users in the company.
The Notes UI was …..cryptic at times, supporting it was challenging – but the features it had at the time for those that were interested were amazing – especially for the job we were doing which was engineering change control across multiple sites and countries. Loved 3, quirky but effective, 4 started getting a bit fancy but still good. 5 was getting pretentious – never made it beyond that.
Be interesting to see if anybody out there uses HCL Domino,
Exactly, and 99% of the company just needed a proper e-mail and calendar function. So what was all that “it is so powerful” urban legend around — told by expensive Lotus Notes admins, who never solved an actual problem?
I spent 15 years at a big consultancy (truly big) and the first 10 or so were on Notes. When people talk of Notes, I think most miss the point. What it was really powerful at, in my humble opinion, especially when you would see people flying to places with really bad connectivity was, at the hands of good Domino devs and admins, at async.
We had hundreds of custom apps for data processing, file sharing and data gathering. You could update these apps and they would get updated the next time you opened Notes. If the data structure had not changed, the next time you connected your updates would be synchronized with the rest of the world. Distributed app management in such way was not possible at the time and web apps were not really a thing yet. You had to go for dozens of proprietary software, develop your own, track versions. With Notes, the whole company could run from within the same platform and it really could do wonders at the time. We did inventory tracking, billing (with integrated email distribution functionality), software package management, mobile phone management, everything on a global scale with 20K+ users, supplier management, data gathering, all through Notes on a global scale.
Of course, to Andreas’ point, 99% only needed email, and it was really overkill and hard to administer for most places. Some people truly loved it and we kept offering Notes e-mail client with O365 backend for a while (yes, lol). What I hear from my former colleagues is that the priority now is SaaS and, when not possible, built-in webapps. User friction is a bit higher. Before everything looked like Notes and now each vendor has its own interface. Administering is also harder because you need to do tons of custom integrations for authentication and authorization (yea, it’s an old business, so there is always legacy stuff around) and data exchange done via APIs that vendors not always offer and sometimes require payment for the vendor to develop it. Before IBM billed us once per year and now there are dozens (if not 100+) vendors to track.
edit: everything looked like Notes, and equally awful, but once people got hold of it, they could use it well.
Excellent point about async capabilities! It’s often overlooked, but it was a critical point for a multinational back in the days when fast, local infrastructure could not be assumed, especially for remote workers or field workers.
It’s the default mail client at the university I work for. It’s use is not a delightful experience.
You put that very kind and nicely.
What stroke me most: literally everyone (i know) hated and cursed it. But then the Lotus Notes admin came and explained how superior and powerful it was and that we only used the e-mail. Which of course was damn right, because all we wanted or needed or cared was proper e-mailing! (And I am not talking Outlook here, though even that one was better than Lotus Notes.)
The user experience is indeed stuck in the 80s/early 90s.
Which version are you using?
There were only two sides in that equation:
1) the Lotus Notes cultists, who loved it and made perpetual money from being irreplaceable once this cancer had been infested the company
2) the actual users, who just hated it. Everything of it. E-mails, Workflows, Forms. (Only calendaring was decent, yet presented in this terrible UI.)
Show me one single user (who was not part of the cult) who ever said: wow, that’s actually a smart solution to my problem and I can work much better now.
As an email client it was OK tbh. I used it for about 5 years until we migrated to Microsoft.
The best element of it (that I still miss) was the quality of the search. It was fast and effective. The rules and filter system was also incredibly powerful. Much more so than even modern Microsoft.
Yes the interface was dated, but functional. It was a genuine surprise how they never gave it a face-lift. I think that would have solved 90% of the complaints about it. It Looked old.
It had no proper threading. No commenting. No advanced attachment handling (like compressing on the fly).
True! But that did not yield much when your mailbox was restricted to a few MB (because Notes DB space was so expensive — at least that was what the admin kept saying.)
Notes DBs used to be very expensive indeed. The licensing was weird (but I don’t recall the details). When Notes started truly hemorrhaging customers, IBM became very flexible and accommodating. I don’t know how it goes under the new owners.
All your rant about Notes, and insults towards anyone that did not hate it, is based on your own bad experience with it.
Not everyone had a poorly managed Notes/Domino environment, and not everyone was restricted to e-mail and not benefiting from the rest of the package.
Developers loved Notes because it enabled them to do a lot in a single environment, they could easily push a new app to users (RAD), interface with any DB and business software, schedule business processes and so on.
Architects could deploy servers close to their users in every company office, and let data synchronize in the background at night, making the best use of the low-bandwidth interconnects.
While the mail handling was odd in many ways, the mailbox template could be customized to meet business needs and processes as well.
Many companies benefited from this all in one software (e-mail, apps, information management, batch scheduling,..), and struggled to replace it when they had to move on.
Also I don’t know where this expensive space stance comes from.
At least in our licensing plan, it was per-user licensing.
Space was as cheap as buying new disks. I just remember some users had mailboxes much larger than what Office 365 could offer at the time we switched (50GB).
Hu? Where did I insult anyone (in this thread)?
Trust, me: if I wanted to insult, you surely would feel it. Although there was no need in this case: Lotus Notes is one of the most hated pieces of software I have met in my life (after SAP maybe). I have no polemic to contribute. Where I come from we don’t kick when a man is down on the ground already.
Well, calling “cultists” people who don’t hate Notes as much as you do feels like insulting them to me.
You said that you only used Notes for e-mail, while it came with templates for many kind of apps (news/blog, discussion forum, library, teamroom, …) that you seemingly did not explore
You said that Notes does not support e-mail threading, while it actually does. At least it did some 15 years ago, last time I used it.
You don’t seem to be in the best position to give definitive statements about this product.
Notes/Domino had a bunch of oddities that did not make it popular.
Localized keyboard shortcuts was pretty annoying for example: Ctrl-F would close the current window on the French client software (F as in “fermer”) whenever you wanted to “find” something instead.
But it also had its own killer features, so let’s add a few others to what has already been mentioned here.
Its “formula language”, which is kind of a macro type of language. It just feels like magic when used to agregate/transform/consolidate data on the fly in views, or in simple scheduled agents, and works with unmatched efficiency.
Sametime. Notes came with its own instant messaging/conferencing app, way before we heard of any Skype for Business, now Teams, on the Microsoft side, and you could either use it as a basic integrated feature in Notes itself, or as a full-featured standalone client app.
And yes many users missed most of the product’s value because the local IT did not manage it properly (and IBM did not help in this regard).
But many companies also ran a large part of their information system on this product alone.
And most where somewhere in between.
Dear Andreas, you are a very smart guy. Could tell me what kind of distributed software solutions you would suggest from the the 90s which could provide easy to develop applications (without recompilation and with built-in NoSQL database solution) that supports both our on-field workers without internet connection and our office workers with connection. The package should come with automated deployements and out-of-the box e-mail integration. Of course the on-field workers application should synchronize with the office workers whenever connected to the internet whithout any additional development on our part.
Thank you, I hear this often :-D.
But let me return the favor: Why would I be interested in such a chimera when all I ever needed (like 99% of the end users) was e-mail and group calendar? Are you the one showing up with a hydraulic excavator (and the 999 page manual) when you want to connect your Garden Terrace with the houses electricity with a simple extension cord?
Your narrative reminds me of our Notes Super Admin, who always told me why simple things could not work because Notes was meant to do so much bigger things.
And remember: the 90ies was the time, when assistants printed every e-mail out, signed and stamped the date before putting it into the executives mail box (not the electronic one).
Andreas, you are interested because you have a real business with sales people, a whole sales and marketing team and you provide around the clock support with field repairmen who will show up at client locations.
As a CEO, you might be fine spending the day drinking coffee and reading e-mail and you can check the calendar time to time which is the day when the dividends are distributed, but when there is an issue in the hydraulic excavator factory those who actually do the heavy lifting for your company will be there servicing the assembly line. Now it is better to have the whole parts database available on their laptops because it is 1999 and in that remote part of Nevada it is still 14.4kbps transmission where it tops and when they fax over an order form the team will need to handle the workflow somewhere.
So this is where Notes was heavily used.
Fast forward to 2005 the situation was somewhat different. We had Firefox with 8% marketshare, Internet coverage was better (still no ADLS in that part of Nevada though), but the tech was still not there to replace the type of industrial applications we used Notes for. It was around 2008 when Chrome was relased with multithreaded tabs (do people remember how it was before multithreaded tabs?) that finally the technology was ready to take over Notes. Now I could say the user “I know the form is broken, but we are working on Jenkins to Hudson migration for all our apps so it has to wait”, or I could say “yes, that new report is a great idea, but instead we will work on single-sign-on/OAUTH/IAM/whatever fashionoble login problem you have”, or I could discuss “sure, we could expand the list of products our application supports, but we are in the middle of migrating to this new UI framework, but let’s return to this next quarter”
I am happy when you liked Lotus Notes and it matched you expectations. But you will have a hard time to convince me (especially when I as a banker really just needed e-mail and groupware).
At the same time, in practice it was not as omnipotent as you seem to remember:
I might not be as smart as you, but for a layman this reads like “it tried a lot and was not really good at anything” — and this matches my own personal experience. Goodness, I start to praise Exchange Server when looking at Lotus Notes — and I hold a deep grudge against Microsoft.
Let me address your followup here as OS News does not allow more nested comments (actually it should not have any nesting but that’s a different story).
You are the one who tries to paint things one sided . I never mentioned that the criticism of Notes was invalid, it is you who tries to ignore why it was once popular how the context changed around it.
I don’t think that it is good to look at older tech with ignorance, why C was modern and popular we already had things like garbage collection and byte code and metaprogramming, yet it took 15 years for these become popular concepts. Similarly NoSQL databases were a thing until they were rediscovered in the 2010s. I could argue that xml configuration in Spring was a mistake, but I can’t skim over the fact that xml configuration did not require compilation or unit test runs which is an extremely cool thing.
Lotus Notes had many good things and many bad things, and before Web2.0 it was more good and bad. Then Web2.0 happened and the technology became outdated. So did Flash. And Struts. And jQuery. And WPF, we had Tapastry, and JSP and JSF, the mighty enterprise Java beans. Then we had AngularJS, and all those other javascript frameworks. Scala became outdated faster than to become popular. Haskell never made it. Clojure, Ruby-on-Rails, Django came and gone. That’s the nature of technology.
Dated, horrible UI – but the strenght of it was not in the client or the mail management (was using Eudora back then…). It was the Sync feature, which added with some part of application building allowed it to be the darling of some IT departments – think sharing data between multiple offices.
Last (potential) customer we talked to move their applications (not the email, alredy migrated) from Domino / Notes to something else was 5 years ago.
Just to say.
yup.
I used Notes for 15+ years. As an app platform, it was quite good, even if the client got fatter with each version. As an Email/Groupware thing, it was always years behind Outlook. Naturally, the last thing to be phased out at my place of work was the Email functionality 🙂
Two companies I worked for in the 90s were “Notes shops”. Back then, at least in the US, the dominate player for office-email was Notes. Outlook (or Exchange as it was generally referred to back then) was still sort of getting up and running. I had come from a Unix background, so I wanted email in a terminal window, so any of these things were new.
Looking back at that time period, I think that Notes was mostly misunderstood. I’m not saying it was great, but I think even Lotus didn’t really understand what it was supposed to be. Notes is largely remembered as a corporate email program, but it wasn’t even initially built for that. Notes was a business application platform. The thing we do now with web-based apps and frameworks. The two companies I worked for had entire company-specific processes and tools built inside Notes. Opening tickets with IT, tracking facilities inventory, and so on. Email was but one of those tools. Lotus provided email in Notes almost as an example of the types of things that could be built in it.
The largest Notes places I’ve seen have huge custom built inventory systems or some sort of database system built inside Notes and migrating off of that is obviously very expensive, so as long as the lower cost is to continue paying whatever company now owns it to keep it running, well that seems like the easier decision.
Notes continued to be picked in companies when a cross-platform “thing” like Notes was needed. You had your engineering department on Unix workstations, back office corporate people on Windows, and the art and marketing departments on Macs…..Microsoft was 100% uninterested in catering to any of that, but Notes was cross platform. (to a degree, and that started to taper off as the 90s marched forward).
Now as for email, Notes was absolutely terrible. Totally non-standards compliant and the way it handled filtering, search, even replying was just bone-headed. It’s unfortunate the Lotus Notes Sucks site went down, but maybe that’s because it was hosted in Domino. 🙂
When I was just out of college I interviewed for a development position at Lotus, for Notes. The manager I interviewed with said that speed was the key differentiator and so the plan was to implement Notes on every platform in assembly language. I was struck dumb at that point and lost interest in the position right there.
BB in my pocket, BES+Domino, Notes on a company-issued (IBM!) ThinkPad… At the same time, also big iron Solaris machines, Sun Studio…. Oh, halcyon days… Also f*k you FlexLM.
It is interesting nobody brought up the young zeal to rewrite existing applications with new, shiny languages, with Rust being the largest elephant in the room.
And those resets, while bringing in real, tangible benefits, also bring in all the bugs that have been solved over decades.
Somehow we have run into the “survivorship bias”, and when every other kind of software bug is rare, people think the remaining ones are the most important. (Hint: a logic bug like leaking passwords as plaintext is more important than occasional buffer overflow. And yes this happened)
I usually bring up the timeless article from Joel Spolsky
https://www.joelonsoftware.com/2000/04/06/things-you-should-never-do-part-i/
[ Edit: I have realized he removed rest of the series from his blog. And the Office relevant part… JustSystems and Ichitaro software, which was a Japanese product which fell into this trap, went from dominating their local market, to being irrelevant due to a rewrite.
Unfortunately for us that story is no longer up ]
sukru,
It’s interesting for someone who dislikes rust to have brought it up 🙂
The problem with the “let’s leave it in C” stance is that it stems from a false confidence in the quality of old C code. C devs often put faith in humans writing correct code despite the language’s notoriety for faulty code. They want to assume everything has been sufficiently reviewed by humans to mitigate C’s faults. The problem is we keep uncovering faults that would have been prevented in rust and I don’t think this will ever stop. Our track record as humans is poor when it comes to fault prevention in complex code.
“The Linux Kernel is Falling Apart.”
https://www.youtube.com/watch?v=BxdWlV5LI_4
Obviously developers are very stubborn. Those of us who have always used C aren’t quick to change, faults be damned. However there’s no denying this harms the industry through never ending software faults, a large portion of which don’t occur in safe languages. We actually can solve a large number of these classes of faults for good, but we have to replace the language responsible for them. The crux is that I don’t actually believe that older generations are ready to change their ways. Widespread change won’t happen until older generations retire, making way new generations who are much more amenable to safe languages.
Or just have taste: the Rust syntax is really ugly in my opinion. I fully buy into all the technical arguments, but THAT syntax!
Andreas Reichel,
I think this has come up before. I agree with you….while rust’s features are very useful for creating more robust software, I find the choice of syntax unnecessarily foreign and regrettably off-putting. The language resembles older academic languages, which might make sense in a way but IMHO has had the unintended consequence of increasing resistance to it. Using a more familiar syntax could have improved adoption. But alas…they delivered what they delivered.
Of course there’s lots of safe languages that use garbage collection, to this end I like dlang and C# personally. But Rust reached an important milestone in solving the safety problem at compile time without a runtime VM. Rust are not the only ones on the block, Ada SPARK took their own shot at a compile time safe language. For better or worse though Ada doesn’t get much attention in industry. IIRC Ada’s claim to fame was that government contracts used to require it, although I don’t believe this is the case any longer.
Absolutely and really nobody can argue that. But emotions always beat reason unfortunately.
As far as I know this is still holding for the defense industry (because Ada can be proven correct). But I am a blind talking of color here.
Andreas Reichel,
Completely off topic…
I’m really curious if there are other types of senses that are trivial to understand when you have them, but basically impossible when you don’t. If you had vision and became blind, you can retain your understanding of what vision is like. But it must be really strange for people who never had vision. How can they even imagine what real vision is like?
This lack of natural senses could be why relativity and quantum mechanics are so counter-intuitive. To a theoretical being that evolved a biological ability to experience this phenomenon, these concepts could literally be child’s play and their mastery of these advanced physics concepts might be taken for granted.
It seems a bit sci-fy to conceive of a being that’s so big or so small that they experience parts of physics that we cannot. Perhaps they’d be aware of entirely new branches of physics that we haven’t discovered yet. I’m just rambling, but these are the kinds of questions I enjoying thinking about.
I assume, you can apply this to anyone “on the spectrum”, example: home-sickness — I have no grasp of this concept and literally can stay anywhere, as long as it is comfortable. Patriotism/Tribalism, same. I hold a passport but except of easy travel my nationality or race bears no meaning to me. Religion is another one: I have no idea how to believe in spiritual entities. In fact I find it easier to explain color to a blind person as a physical phenomenon than any of those 3 examples.
Alfman,
I don’t have a strong opinion on Rust. The syntax is not for me, but it is overall a good language.
But, what I have the issue with is trying to force rewrites even when it is a net negative sum.
That is true, but it comes with a cost. When someone replaces a software with 20 years of heritage, they also have to re-learn many of the lessons in those two decades, or they really leak passwords plaintext
(I’m not joking, Ubuntu had this in their new and shiny re-implementation)
“Green field” is possible. But is a much harder undertaking than most people realize.
It should at least start with a very good “acceptance test suite” for example
(Java has one, and that is how many different implementations were able to keep compatibility)
sukru,
Yeah, this is a common opinion.
I disagree. While there are short term costs in replacing a crooked foundation, there are also long term costs in keeping that foundation. Over time it’s obvious that the long term costs that will dominate. C keeps contributing to decades of software faults and if we want more robust software we’ll need to move on. It’s hard to convince people to pay to replace problematic software for dividends that pay off in the future. Alas, this is why I predict little progress on the software robustness front until generation “C” retires. Haha.
Indeed, a replacement takes a lot of work.
Aside: AI might actually be able to write replacements while minimizing human work. The technology may not be ready yet, but I think most of us are going to live to see the day when this possibility turns into a reality.
I like having the opportunity to fix past mistakes, but in reality we are frequently locked into them. True green field projects are rare in software. It’s easy to point out the ways technology has evolved suboptimally (we should have done X, Y, Z), but it’s much harder to break people’s dependency on that technology.
I don’t think, that the language matters much (although I would prefer Rust over C when I was not just a dumb Java developer). For example, I stuck to plain Java (and a bit of Groovy for scripting inside my applications) but never jumped onto the Kotlin or Scala wagon.
However, UI frame work matters. A lot. And this UI frame work (via available bindings and also paradigm) very much decide about the language as well. I never moved to Rust because the preferable UI frame work Iced is not mature especially around table widgets.
So, in regards to your rewrite theorem: Yes, rewrites fail (I failed with of my own projects) because a) the old one is good enough (which kills any better) or b) the rewrite uses the same UI (which by the time of the rewrite is completely outdated and the market has moved on) or c) the rewrite uses a new UI which does not cover the complex uses cases that have grown in the old UI (and so people shun it). I don’t think it is ever the language itself (“Rust”).
Andreas Reichel,
Haven’t done much UI migration recently, so can’t speak about it.
But I agree lang8age does not matter much. One should use the right tool for the right job.
I’ve had many dealings with HCL (Hydrabad Consulting Limited) and they are probably one of the worst consultancies out there, I’m not surprised Domino/Notes isnt’ doing well under their stewardship.
Hello
Just a few things:
– Ray Ozzie designed Lotus Notes in the 1980s
– I used Lotus Notes till version 9 if memory does not fail on me.
– At that time (Lotus Notes 9) Domino was the server, Notes was the client.
– IBM decided to change the “Notes Client” to eclipse.
– IBM ported OpenOffice to eclipse to merge it with Notes 8. It was called Lotus Symphony (the second one)
– Lotus Domino/Notes was not only an email platform. It was a documental application server and development tool. The development tool was called Lotus Designer and uses a language called LotusScript. IBM had lots and lots of Domino apps for it’s internal use.
– The Notes community wanted IBM to release the .nsf format as open source, but it never happened. Instead of that, IBM released Lotus Designer for free with some limitations (I guess). I call the Lotus Manager of that time “The Guy that Killed Lotus Notes”.
For me, Lotus Notes/Domino provided the dream to the casual developer, an easy way to create applications, like VisualFox, VisualBasic. But since it was proprietary it had the same fate as those technologies.
Regards
Martin