I like Excel, the only issue I have is with people using it as a feckin' database instead of using an actual DB. Having said that Access was\is horrible so I can see why this happened.
I remember Lotus 1-2-3 very well. It really was as widely used as all the history-of-Lotus stories claim it was. In fact, back around, say, 1984, when almost no software package had a monopoly, Lotus already had its particular niche locked up tightly. That’s right: WordPerfect was still a serious competitor to Microsoft Word …
"using it as a feckin' database "
You obviously haven't encountered people using Word tables in multiple documents as a relational database.
People learn one tool and try to apply it everywhere. Get training? No, a computer science education should be in the help file shouldn't it???
Yes - and in the late 1980s, every accountant I met wrote all their letters and their CVs in Lotus123. Not a single one them used a wordprocessor. I don't anyone who took MS Word or Excel seriously for years - the MS products got their chance when the incumbents screwed up moves to 32 bit.
Sadly, I've actually saw my mom write a CV for my dad using 1-2-3 a long time ago. Yes, we do have a Word Processor program with that old 8086 clone (I believe it was called PFS Professional Write), but she just sees 1-2-3 as easier to use because her company actually sent her in for training in 1-2-3 and nothing else.
You obviously haven't encountered people using Word tables in multiple documents as a relational database.
Geh, you're gonna bring back my nightmares. I was given one of those things that clocked in at 800 pages a few years back and told to 'fix' it. Yeah, they really do exist and yes they're every bit as bad as you'd expect them to be.
For the record, my fix ended up being a script that read the word docs into a proper database (MSSQL to be exact).
I like Excel, the only issue I have is with people using it as a feckin' database instead of using an actual DB.
Well said. Excel is great, but for some reason people turn it into a word processor, data base, project management chart, and much more bizarre home-grown applications rather than just using the right tool in the first place.
I work in a bank's IT department, I used to work on an application doing massive amounts of recursive compound interest calculations. We had actuaries produce formulas for the calculations and had to take the upmost care in coding it to take account of number formats, decimal places, rounding, floating point approximations, etc. as it would all massively affect the results.
The testing department compared it all into an excel spreadsheet, knocked up by some monkey over a lunch hour, and refused to accept that when the output didn't match that the application could actually be right, and that their spreadsheet might be the issue.
No word of a lie, I have seen someone use it to produce building layouts. The mind boggles.
I used OpenOffice Calc to keep track of users, their desks, port numbers, coordinate location, date last seen etc.
I know that's delving into DB territory, but it made it really easy to implement, and coexist with the cartography software I was using to map the desk/floor layout.
And yes, I know that was overkill, but the entire process streamlined quite nicely together and produced a very usable product to find any person, on any phone extention, on any port, instantly.
And most importantly, it cost little in time, and cost no additional money - which kept the boss happy.
On the other hand, we had to support staff that were asking how to overcome the column limit in Excel because their yearly finance spreadsheet had run out of space for their years...
re "using Excel as a Database" . I completely agree . It is a crime against computanity , that I must confess I was once guilty of , with VBA macros to glue sheets together coz 65K lines wasn't enough. Please forgive, but also please down vote this post so that I can truly atone for my sins. I have sinned , but I have seen the light ! (beyond Excel Databases, which I don't suppose really counts for much, but it is a start )
They did exactly the same thing with with 1-2-3 !
There were spreadsheets with macros you could load that made 1-2-3 into a word processor - a rather poor word processor, more like a fancy notepad (lowercase n, it is not worthy of an uppercase n). Fixed width at 80 characters to match the printers at the time.
I also saw one spreadsheet that made it into a primitive project management tool with very simple Gannt charts.
"The sad thing is that 20 years later Excel *STILL* can't cope with cells that contain dates or times and insists on formatting them into some monstrosity that in no way resembles the actual contents of the cell."
Eh? I've used dates and times in cells in excel loads over the years ...
Perhaps he's referring to the irritating habit of Excel formatting DateTime values from a SQL Server result set as just hh:mm:ss.
It's something I see often; you run a query, and want to quickly play about with it in Excel: Select All, Copy with Headers, into Excel, Paste, and every DateTime value from SQL is formatted as "Custom". You have to select the cells and change from the "Custom" format Excel has decided upon to a Long or Short Date or whatever.
The problem is the other way: Excel corrupts data by autoformating something it believes is a date (DEC12, for example), when it is not a date. There are other examples of issues while pasting data as well.
remedies for that are nontrivial, as it isn't a setting you can turn off. The only fix is to preprocess your data before importing.
Olader versions of access used to have some amusing americacentric ways of dealing with dates but I can't remember having that issue with excel.
Access isn't awesome, but it can do some serious work if you stick within its intended scope. Despite having oracle to play with, access was the easiest way to quickly put together a project before lunch. It was shocking if you wanted to support 80 people, but when you worked in an environment where you would need to finish 2-3 different projects a day modelling data from customer accounts it did the job well, on pretty standard desktops for not that much money. It may have been 'dirty' but it could be pretty quick. I'm probably biased though :-) I think people expected it to be able to be oracle+forms etc all in a neat package when in reality it's a baby database with a decent ide that if you are desperate you can tie into a real db backend.
Excel dates from a 3rd party data spreadsheet used as an import job.
If the cell is date format, it imports as something like 127994.
The cell needs to be in text format, which obviously cannot validate dates, which leads to things like dd-mm-yy vs dd/mm/yyyy or the day/month/year seperator left out.
"It's even sadder that that is trivial to address and you haven't worked out how in 20 years."
I'll bite, troll.
What's the "trivial" way to fix it so that Excel never attempts to auto-guess that a cell is a date or time (because it always gets it wrong) when double-click a csv file (say, one attached to an email) to open it?
And don't say "save the file out of the email, open excel, file -> open, select file" because that isn't trivial when you're dealing with numerous auto-generated database exports on a daily basis.
P.S. 6 downvotes? I'm guessing most of you aren't actually using Excel as a quick way to pass around adhoc database extracts.
Last time I had to do this, I used a perl script to manage the .csv's into whatever the right way to format a date value was so that excel could recognize it.
Not sure if that would work for you though.
This was 5 orso years ago, output to .csv -> perl script -> email -> excel 2003.
Took some fiddling, but after that the script worked fine.
Not sure what became of that script though. Couldnt find it , and cannot remember what the exact magic format was that made ezxcel get it right. Maybe just simply yyyy-mm-dd hh:mm:ss
This post has been deleted by its author
"I like Excel, the only issue I have is with people using it as a feckin' database instead of using an actual DB."
Well, I don't fully agree there because if a program provides a certain functionality then why shouldn't people use it?
But even so, I don't think you should blame the users over this but Microsoft instead. After all; for many years now a standard Office package consisted of Word, Excel and PowerPoint. SO what do you suggest Office users should use for a database instead?
I can't really think of any usable and logical alternatives to be honest.
And as I said, my personal opinion on this matter is "if it works...".
"But even so, I don't think you should blame the users over this but Microsoft instead. After all; for many years now a standard Office package consisted of Word, Excel and PowerPoint. SO what do you suggest Office users should use for a database instead?"
That would probably be the case yes. But you also find people using Word tables for this, not to mention a "presentation" made up of images on Word pages - or worse, and image per Excel tab. Now why didn't they go open PowerPoint?
As for me, I used 123 in the 80's ... for financial / engineering calculations. Yep, "what it was meant for". If I wanted to write a letter / specification / manual I started PFS:Write. For database I used R:Base and later for smaller DB's Q&A. But "upgraded" to Ability+ in the late 80's - combined all those into one (like Symphony should have been, only much earlier).
When the MS-GUI abominations happened with the slap-on Win3/3.1. I looked at it a few times and decided to give it a skip. W95 wasn't much better, but had a better UI and finally your programs could actually operate. Win NT 3.5 at least didn't crash on you (all the time) - put programs tended to "not work" on it if it wasn't written specifically for the platform. When NT 4 came out, that was it - MS should've stopped there, it only got worse after that.
BTW, in the mid 90's I used Excel/Word/PowerPoint/Access or that stupid Ms Works (which was a contradiction in terms) at university because that's what they gave us to use. At home I used Borland Office (including WordPerfect / Quattro Pro / Paradox) later editions also included Presentations. It was cheaper than MSO and had more capabilities without the hoops you had to jump through. Not to mention Paradox at the time was ahead of Access - and was highly integrated into their full DB InterBase if you wanted to work with large sets & a lot of concurrency. Quattro Pro was also a lot nicer and more fully designed than Excel, never had an issue with any date formatting / queries / CSV import-export. Used to look at Exceller's with a blank stare when they complained about this, I simply did the "real" work in QP.
But as the article states, MS's programs might not have been the "best" for the job at hand. They were the "best" marketed - be that some backhanded way or not. So they ended up on nearly every desktop there was. No wonder DOC/XLS is still the defacto standard when people send files to each other.
There is one more problem with Excel, when using as CAS, or a stat tool. In particular, when relying on it too much with simulations and actuary analysis, which is unfortunately a pretty common practice.
In many cases it was shown to not calculate reliably. Unless MS has fixed that already. This is one of the paper to talk about some issues. Calculations done in MS Excel can be much less accurate than using Gnumeric (best of spreadsheets really utilizing R libraries), or even l(o)ocalc. For spreadsheet stuff, personally I'd go with the Emacs org-mode. Of course, proper CAS is to be used instead.
Some students of mine that once tried Excel for projects got a few things messed up with a chi-squared test, and few distributions, such as Bernoulli and even the Normal.
Completely agree, Excel is an accountants tool, and most accountants use bespoke databases or accountancy packages.
If you want to build a mathematical, scientific or engineering model, there are much better software like math-lab or my favorite was TKSolver.
For forms (something most people use word and excel) Infopath works really well (I have yet to find an alternative).
For a database, I liked Access 2003 to quickly build and work with data. Since Access 2007 though it has become a bit more difficult. The world could do with a simple database program for quick works with data, that is as easy to enter info as excel, but has the function of rational database. Naturally if you are talking about a multi-user system you really should be looking at a database server and a proper developer.
It is I.T's fault. You Microsoft fan boys take a user who says "I need to collect some info" from a one day LAMP web site form and turn it into a .net $100,000.00 fiasco. Of course the users are going to tell you to feck off and then resort to simply emailing excel sheets to people to "fill out".
"Lotus got complacent. They stopped innovating, or fiddled while Rome burned, or made stupid decisions. "
Not just complacent but the head of the company was IIRC the highest paid executive in corporate history while these stupid decisions were being made. The US$20M he was taking home as salary in the 80s would probably in one year be worth more than the max $200K pa that Bill Gates got as salary throughout his entire tenure. (I'm not counting stock options for either, just salary).
Not long after Ashton-Tate produced the enormous fail that was dbase IV, and Borland began funding Philippe Kahn to make vanity music discs. Not hard to see why I ended up skipping from Lotus 1-2-3 to dBase III to Paradox/Pascal and then on to Microsoft products. I loved each and every one of those other products until their managements started competing in a shark -jumping olympiad.
@hitmouse So very true. Complacency every time - methinks its something to do with CEO's (once they hit the big time) surrounding themselves with sycophantic 'Boards who keep telling them how wonderful their (completely irrelevant) product/thinking was.
You could add WordStar to this time frame, just about ceding top slot to WordPerfect (Word was generally laughed at).
Yup. WordPerfect, Lotus & Dbase - top of the heap. In some cases worse, in some better but in every case the emerging product became the easier to use - a salutary lesson to people today who think their product will remain top dog because it's better engineered - ask Nokia
The large company I was once a part of tried standardizing across all sites on certain office programs.
To that end, the standard was set as MS Word for documents, Lotus 1-2-3 for spreadsheets and Groupwise for mail (and also I think whatever Lotus's then presentation offering was)
As our site was engaged in research work, we liked to make lots of graphs to show off our findings. And also use Excel's analytical add-in as well. WE did try - honestly - but unless it was a bar chart or a pie chart 1-2-3 was rubbish for graphing.
Upshot was that the site more or less got a blanket exemption from the diktat and Excel remained on the desktops.
I remember this particular period well, as I worked in IT training at the time.
All of the big boys who made the industry standards at the time - 1-2-3, WordPerfect and Paradox (ok, not a standard as such but still immensely popular) suffered greatly when porting from DOS to Windows. Anyone else remember WP5.2 for Windows? Yeuch.
By the time they'd got the glitches ironed out in basic functionality it was too late.
Borland pretty well ruined Paradox, and dBASE IV as well. But, the lack of backward compatibility was something Microsoft did with Access as well. I recall being pretty well stuck in Israel and screaming at an MS support person when they said, "oh, you can't do that any more." I wanted to carry over unchanged values between records. Turned out you really can do it, but, it was no longer a trivial bit of code. In the first iteration or two the process was as simple as it was in dBASE. But with the third(? - around 2000) iteration the company decided to "help" all MS users by switching over to VB without fully implementing the capabilities of the system it was replacing. Paradox was and still is in many ways vastly superior, and for coding and quick apps, the dBASE-based languages (dBASE, Foxpro, Quicksilver, ..., were all way better).
Excel gave Lotus 1-2-3 the kick up the backside it needed - a few years later, 1-2-3 was back to being far superior. However, by then it was the minority... I still have a copy of one of the last versions running on my desktop - it has many features that are even now more advanced than Excel! Just a shame it's missing some of the more basic things that we've come to expect from Office...
I used to do "1-2-3 programming", writing macros for clients - we used to write "real" software as well, but people often wanted automated worksheets.
One set of macros was so complex (it used to download sales forecasting information for a major chemical company from a VAX and produce manufacturing plans), that it started doing "random" things. If you stepped through in debug, it worked, run the macro and it would do different things every time!
In the end, we contacted Lotus. Their reaction? Wow, we never expected anybody to do anything so complex with 1-2-3. :-D
There was nothing unusual about the way Lotus disappeared. Yes it was an excellent product and no, it did not have much competition, so it is not a surprise that it thrived. However, as soon as something better came along, it had the choice, adapt, or die. From the same era, consider DBase III. Clever name aside, it was the dominant product in the market until something better came along in the form of FoxBase. Neither product survived the transition to modern GUIs.
Er, Foxbase got bought by MS. Hung around for a while with a GUI tacked on. One unfortunate feature was that hitting the "Close Window" icon while in the Fox Read Cycle left the thing hung like a dog and it was reboot time(!)
The engine ended up as the JET engine in Access, which is why MS bought the thing in the first place.
"The engine ended up as the JET engine in Access, which is why MS bought the thing in the first place."
It was Fox's Rushmore query optimization that got added to Jet, which already existed when the two companies merged in 1992. There was probably a lot of code merging after that while ODBC was being created.
As I said, the drawback was that if you thumped the "close" box (red "X") on a Fox app window, it left the old skool DOS engine sat in the background, twiddling its thumbs and waiting for input that would never come. Insult heaped upon injury was that you couldn't disable the close box as the OS put that in.
I remember doing a training course in which this was pointed out and demonstrated. I noticed that one of the MS demo apps actually shut down gracefully when this was done. The trainer opened the thing to look at the code and swore long and hard. There was a code block helpfully commented to say that it was handling the graceful close........which was both nigh-on rocket science and on a par for size with the rest of the application.
>Wasn't Jet used as the database for Active Directory?
The Exchange mail database eingine is also called Jet, but it's not the same.
The Exchange directory (which was extended to become Active Directory) is also used by Exchange, but it's not the same as either Jet or the Exchange mail database engine.
Jet is a multi-user database engine. Exchange is a single-user database engine. AD is a replicating database enigne.
...it was the dominant product in the market until something better came along in the form of FoxBase. Neither product survived the transition to modern GUIs.
Not through lack of trying. Just in last few years, I had come across an application that used a Fox backend to handle the database side of things.
We were trying to get it off one old machine, onto a newer box, newer OS, and in the process of trying to get the thing to work, I don't think I swore that much ever before.
Re access. It's not a great database for proper multi-user database applications, but it's a bloody wonderful RAD solution for the horrors you find lurking in Excel.
RE 1-2-3 I wote macros for it and remember they had a few decent features in later versions but nope they were beatend fair and square. Funnily enough the other obvious place a MS product whacks the compeditor is Outlook vs Notes. (The Notes client being clearly designed to make the user pay for sins in a prior life).
In one place I worked at the company fax template was an Excel document.
I even created a Word template that looked identical but no, the management knew how to use Excel.*
* They didn't, the wasted hours I spent fixing up broken fax docu...sorry...spreadsheets.
Then there was the manager who named every one of his spreadsheets with variations on his own name regardless of function. Not DaveFinanceReport.xls but Dave.xls, Dave1.xls David.xls etc.
This post has been deleted by its author
I'd say that all places I've worked at where Excel ( and Access/IE6 apps) are used in any anger are seriously limited by them.They are all mission critical bits and pieces that were bodged together in such a way as to be almost completely unmaintainable.
They put you on the ladder to programming heaven but then you discover you cant get the ladder into the lift that even simple software management offers. They are almost physical barriers for turning an SOHO into SME and SME's into E's.
Excel is a bloody good spreadsheet, yes, and has good graphing capabilities.
But once it had 123 and quattro beat it has more or less stood still (Pivot tables? maybe, though they aren't what my maths tutor said pivot tables were)
Calc and Gnumeric are just clones, no-one is breaking new ground.
maths went matlab or Octrave or mathcad, and the proles were left in the 1980s.
For a few years - late 80s, early 90s - Quattro was better than Excel imho, but eventually Excel got an edge and stayed ahead when it added VBA (there was no Quattro equivalent at the time) and Quattro became an also-ran. Although it is still available after being bought by - oh can't remember who, a graphics company.
Never underestimate what VisiCalc did for the PC world - it meant people with just average maths skills could do projections and try a variety of variations without asking the programming department and waiting two or three days for an answer. It kickstarted the concept of a computer on everyone's desk. No more mystical command line /* /a />9.3 /qr rubbish, just straight basic maths and up popped the answer (if you got the brackets in the right places).
Actually, it became Quantrix. Download a 30 day eval copy and blow your mind, just by following the really good intro doc that comes with it. THIS is how spreadsheets should be, tables with sensible column names like Item, ItemPrice, Quantity and Subtotal, and rules like Subtotal = Quantity * ItemPrice, Total = Sum(Subtotal).
I'm not sad Excel beat 1-2-3. But Improv should have at least made people aware that there is a better way than 'stretching' elementwise formulae over ranges.
Honestly, if you have 10 mins, and you like this sort of thing, check out Quantrix.com. Unfortunately I could never persuade my company to buy it.
That'll be the one.
They shipped it and by all accounts it was jolly good. Shame nobody bought it, but Lotus had built their empire courtesy of the IBM PC and took the IBM line on what was the GUI OS of the future......oops.
It didn't help that running 123v3 (the three-dimesional one) for DOS under Windows was a right, royal PITA as 123s internal extended memory manager for accessing memory above 640k (thanks Bill!) refused to play nice with Windows. Canning your Windows session and dropping to DOS was the only approach, if not having the whole shebang hang like a professional twatdangler at the drop of a hat was anywhere near important to you.
Moving to Excel meant being able to use a spreadsheet without having to close everything else down first.
The director of the accounts department preferred 123 for Windows as he already knew all the keyboard shortcuts/macros, but the office suite was significantly cheaper to roll out across the company - pretty much the price of either a 123 or word perfect licence, so that was what swung it in our firm rather than all this airy fairy functionality and ease of use stuff which we'd like to think is so important.
15+ years later I still would rather use Threads than Outlook though, it was so much easier to follow a group discussion.
15+ years later I still would rather use Threads than Outlook though, it was so much easier to follow a group discussion.
It is entirely the fault of Outhouse that almost everyone who uses EMail (which is almost everyone) thinks that the correct way to refer to a message when replying to it is to include a complete copy of that message (and all the messages to which it is a reply) in that reply -- it's as though RFC822 had never been written. The amount of wasted bandwidth and file storage that this imbecilic duplication has caused around the world is colossal.
I never used threads ... but it can't possibly be as bad as Outlook.
Far too many years ago (somewhere around 1992) I was taught how to use 1-2-3 on a college course (along with WordPerfect 5.1 for DOS and something else which completely escapes me now). Learning all the keyboard commands was pretty painful, but it seemed amazing to me at the time having never even heard the word spreadsheet before. I vaguely remember the Windows version I used in the real world, but moving to Excel was a bit of a revelation but I hated switching over. Took me ages to learn it too, particularly the more advanced features.
The application market seemed to move pretty quickly back then from what I remember, no sooner had you got used to something then the industry standard changed. Everyone used WordPerfect pretty much then suddenly switched to AmiPro (remember the hidden game?), which was fairly quickly replaced with Word. We saw the same thing in the browser wars (Netscape anyone?) and search engines too.
These days there doesn't feel to be much that's new really or that is a real gamechanger because of the user experience or a particular function. Shame really but I'm very glad that I saw it happen all those years ago.
If we fast forward to today: MS are adding HPC extensions to Excel, in order that the said same traders (and similar) who want to make more and more complex calculations can have access to clusters of HPC nodes directly from the spreadsheet on their desktop. People who really "know" about HPC are saying that this is terrible and there is no substitute for proper programing, people who "know" about customer service are saying that it's probably the most democritising thing that's happened in IT for a good decade or so.
My point? If the likes of LO and HPC vendors don't sort out their products, they're going to be left behind, obviously the big HPC clusters are always going to be bespoke specialist affairs, but the small and medium could be cleaned up by MS.
Excel was so successful that the financial industry has built entire applications around it that unfortunately may never go away. Global banking runs on Excel! What's worse MS may never even face competition due to such strong vendor lock-in from Excel customisations / VBA / Macros used for business apps and traders spreadsheets etc.
It would be nice to have alternatives in the market as commented in other threads this week regarding a switch to Open Office / Libre Office. But as its been pointed out there are always Formatting incompatibles and that's just the tip of the iceberg...
Its 2013 and yet there's no Office alternative with FUNCTIONALITY that competes with Office head on! Business customisations in Excel have been around for 20+ years, and VBA in its current form for at least 15 years. So what's taken the competitors so long to do little next to nothing regarding these two key areas?
Its like they deliberately chose not to compete with MS head-on, even Google!!! So why even discuss alternatives like Open-Office / Libre - Office if they don't fully support business customisations and VBA / Macros etc? The entire financial industry in built on these types of bloody spreadsheets and we will never get away from MS as long as Excel is the only option... It is the ultimate vendor lock-in!
That pretty much explains why nobody will challenge Office any time soon. As you rightly said, none of the other office packages support this level of integration. When I'm out and about on my Linux netbook, I might use LibreOffice to knock up a basic spreadsheet or document (and then go take a shower to get clean again!) but if I need to do any serious work, it's all done in Office.
VBA is one of the standouts - With an office app + VBA you can build hugely complex applications with a fraction of the complexity of trying to write a native application using VB and Visual Studio. I just wish Microsoft would do a VBA.NET and expose the .NET framework to Office so you can do some of the advanced stuff like on-the-fly crypto and I/O etc.
When VSTA was being introduced it was at the same time that .Net was being widely promoted by MS. Now that .Net is has been deprecated by MS, does that mean VSTA also becomes redundant?
What can VSTA do that VBA using reference linking to external libraries can't?
Any EXCEL VSTA / VBA developers reading?
"Now that .Net is has been deprecated by MS, does that mean VSTA also becomes redundant?"
Deprecated according to who ?
The latest version of Visual Studio, 2012, has recently been introduced as the de-facto platform to use for Windows 8 development. When installing it first makes sure that you have the latest .NET framework (4.5 at the time of writing) and then proceeds, providing full access to both C#.NET, VB.NET as well as ASP.NET.
Microsoft even provides .NET framework targeting packs specifically for Visual Studio 2012, you can get an overview of those here (link to MSDN site).
Well, tight already mentioned VSTAS and from a personal experience I can also say that you can get quite a bit done with VB.NET (used VS VB.NET 2010 Express for that), although now that I'm evaluating VS 2012 myself I have to admit that there are plenty of advantages to be found there.
But speaking of VBA...
If you want to do crypto or I/O you can, depending on what stuff you have installed. Because Microsoft has provided access to that stuff from within VBA. If you're in the VB editor you can setup references for your project. And that will add functionality to that project.
For example; if you create a reference to the "Groove security context type" library then you'll get quite a bit of encryption support to work with. If you create a reference to, say, the Outlook library then you can suddenly access Outlook right from within Word.
I/O is obviously a bit too broad but depending on what you need chances are high that you'll find a library in that list which you can use. Perhaps the "Microsoft Internet controls" can do what you want..
The article makes the case, though, that Lotus was in a similar position with 1234 at the time, such as the Oracle add-in to 123, and so on, that made it appear to be just as entrenched as current de facto standards are, and yet it still collapsed. We forget how fast changing the technology world really can be. I worked for a financial services company at the time and it's easy to forget how entrenched 123 was.
By the way,t he article fully matches my experience at the time. We held discussions with LDC and MS at the time of the rise of Windows 3, and came away dejected that LDC had no clear way forward while MS not only had good ideas, but understood where they needed to improve. The difference between the companies when you looked at IT strategically at the time was really chalk and cheese.
Sure, everyday banking operations use COBOL and Java in places to do the day to day routine stuff. But a huge layer of analysis is still done using Excel because of its flexibility. For that I think you'd find most banking users have a copy of Excel on their desktops! Investment Banking is actually quite a small area. I'm talking about Central Bankers, Hedge Funds, Pension Funds, Insurances Houses, financial regulators, so its much larger than just Investment Banks.
But Excel... I liked. It is now the only thing I sometimes boot Windows for, as Libre Office outdoes MS in MS-like awfulness.
Come to think of it, I'm not much taken with God, either, so I don't know why I'm telling him this.
Paris, because even she has some decent software
Seems like too many people forget that MS actually screwed the 3rd party people by telling them "Use the API's we document" - then pulling the plug and using undocumented calls in their own products causing the 3rd party software makers with dung on their face because they were not given the proper documentation.
Yes, I was one of the users of that era. I think I had an inkling of the MS dirty tricks strategy.. I was constantly noticing that OLE functionality that made copy and pasting sophisticated and/or formatted data objects from one app to another worked insanely well between MS apps (like taking that fancy looking graph from Excel and plopping it into a Word document, or copying 5 lines from Word and pasting it into Excel), but looked like crap for non-MS apps (bad formatting in pasted content, or no pasting at all).
College students like me at that time knew that something weird was going on.
1. Maybe the non-MS app designers weren't thinking long-term, which made no sense, since the software designers of this time were really good.
2. More likely, the API documentation was sorely lacking.
Every user going from Ami Pro to MS Word would think, "Wow, Word looks like crap!", but if they could work through the pain, the inter-operability made Office the only possible winner.
I well remember the coming of Lotus 1-2-3 and had a similar experience to the author. We had 1-2-3 in our company. I went to a MS show and saw a demo of Excel, and my jaw dropped. I went back to work and told the boss "we need to get this NOW." We ordered it that day. When it came in, I took it home and stayed up to 5:00 AM Friday night playing with it, something I've not done before or since. I was THAT impressed with Excel.
Remember printing on 1-2-3? It was difficult to do. And if you could get it to print what you wanted, it generally looked like crap, like a report typed in EDLIN. As soon as we had Excel, we could do client-presentation-level spreadsheets quite easily and never looked back at 1-2-3.
Well, a PC I got in 1996 came loaded with Lotus Office Suite and I rather liked it, though it was obvious that it needed to be brought up to the state of the art as far as a scripting language was concerned. Unfortunately, I seem to remember IBM acquired the Lotus name around then and killed off everything but Notes and Domino.
I always liked Quattro Pro it was far better than Excel, especially at graphing, and lotus Improv was so far ahead of its time I think if they redid it now it would take off. Well maybe.
Thing is that everyone uses Excel now so you end up sticking with it, but I cannot stand the fact you need to program it in VB to get anything useful done.
re "But this is one of relatively few cases where a software product with essentially no competition simply vaporised. Amazing"
One word: Novell. I entered the IT industry when Novell had over 80% of the server market (even managed to squeeze a CNE out of the company i was working for) and look how it disappeared almost overnight!
Another domination (rightly or wrongly) by M$.
Admittedly I just ran out and got a MCSE at the time (damn easy during NT3.51 days) so I was one of those rats who deserted that Novell Titanic.
I helped a friend install Excel on his new PC. He was coming from a Lisa world. Fun times. Excel was running on top of runtime windows, Windows 2.0. This was my first glimpse into the beast that was coming. At the time 123 was faster and my fingers knew the commands. Mouse? Later I got a copy of Windows 3.0, ouch is a huge understatement.
While I give Microsoft credit for doing a decent job with its Office products overall, starting back in the DOS days, it's not all down to their competitors dropping the ball.
I don't know if any of you remember this, but there was apparently an old adage at Microsoft that ran something like this:
"DOS isn't done until Lotus won't run"
I believe there is substantial credible evidence that Microsoft has historically engaged in some highly questionable tactics to make their application competitors look bad or throw roadblocks at their apps running on the OS platforms Microsoft controls. Particularly back in the DOS days and the early days of Windows when vendors of traditional DOS products had a serious developmental disadvantage compared to a competitor who had intimate knowledge of and resources pertaining to this brand-new platform.
Novell had some particularly damning material that they were planning to use in a lawsuit over exactly that - but Microsoft settled that case with a large, secret payoff to Novell literally the day before the case was scheduled to go to trial. I believe Attachmate (Novell's new owner) is apparently interested in reviving either that case or a similiar one, so these very detailed allegations may yet see the light of day.
Actually, Lotus killed themselves AND Borland, and created the Microsoft monster when they sued Borland over Quatro. The court decision found you can steal anyone elses software note for note, as long as you do it in a clean room, i.e. dont have their code. Same look and feel and functionality ok.
True. And you can partially blame the schools for teaching everyone excel. But I.T. really only has themselves to blame. Users call and say "I need to collect info about our properties" and idiot Microsoft Fanboys in I.T. say "well, meetings, write up a spec, and do a .net web site with a mssql database and web services, (and we can re-write it every 3 months, maybe change to WCF, or silverlight, or...) Pony up $100,000 of budget to start with... So you see, it is ALL the I.T. departments fault. Of course the users say "screw that I'll email an excel sheet to them and have them email it back" Now, if the I.T. guys had said "Mike, make them a LAMP site tonight" it would be a whole different ball game. My point? You ms fanboys in I.T. are the idiots, not the users.
Interesting. This is not quite what I remember. First, IBM and Microsoft convinced Lotus (and also Word Perfect, the other "big kahuna" of the DOS PC era) to spend all their development cash on OS/2 versions of their products. As good partners, both did just that - having ports running on OS/2 by 1989, but without the Windows-style interfaces. Secondly, IBM and Microsoft divorced, followed by the Softies announcing Windows 3 as their only strategy, coupled with their intent to move Excel and Word to that platform exclusively (no OS/2 versions). This left Lotus with a hard choice - stay with IBM and OS/2, or move to Windows 3 and partner with Microsoft, which had the leading competitive product. Having by that point built a working version of Lotus 123 with full OS/2 Presentation Manager (PM) capability, this was the equivalent of telling Lotus "please throw out 3 years of development effort and start over."
Remember that the OS/2 version of Lotus 123 COULD NOT BE EASILY PORTED to Windows. OS/2 software was based on an assumption of a pre-emptive multitasking OS, which Windows 3 was not. Also PM and Windows were quite different on the UI as well. So not only would Lotus have to change the UI over from PM to Windows, it would also have to rework the core code to work under Windows 3 w/o proper multi-tasking. Once it became clear that Windows was going to be the customers' direction, however, Lotus faced the inevitable and built a Windows 3 version, but only after MS had built a big head start convincing customers to migrate to Windows 3 with Excel and Word.
That was when the final blow was applied by Microsoft. By the early 1990's, both Lotus and WordPerfect had working versions of their flagship products running on Windows 3; two years late, but good products. This was when Microsoft announced "Microsoft Office", meaning that they were giving liberal allowances for competitive upgrades from either Lotus or WordPerfect to get both Excel AND Word for Windows. Hard for the competition to match that, given their overall higher price, their need to recover both the OS/2 and the Windows development investments, and that they were by then cash starved because of loss of market share to Microsoft.
Was Lotus "lazy"? No. Did it put too much trust in its partners IBM and Microsoft - decide for yourself. Do I blame Microsoft? Of course not - taking strategic advantage of your partners appears to be how all tech companies operate. Microsoft has just been so much better at it down the years.
I never used 1-2-3, so I can't make a comparison with the Supercalc.
I found Excel more difficult to use than Supercalc. (And Supercalc graphs were MUCH easier).
Of course, Excel allowed Excel add-ins, which made it much more powerful, but long before that it was Supercalc that found the errors in the hand-prepared spreadsheets sent down to us from head office.
Biting the hand that feeds IT © 1998–2021