Concatenate

Concatenate (pronounced kon-kat-on-ate)

(1) In biology, joined together, as if in a chain.

(2) In general use, to link things together; unite in a series or chain.

(3) In computing, the joining together of two or more objects stored in different places; most familiar as the spreadsheet command(s) invoked to join cells.

(4) In formal language, as string concatenation, the operation of joining character strings end-to-end.

1425-1475: From the late Middle English (as a past participle) from the Late Latin concatēnātus, from the perfect passive participle stem of concatēnāre (to link together), the construct being con- (com-) (with, together) + catenare, from catēnō (chain, bind) or catēna (chain) + -ātus (from the Proto-Italic -ātos, from the primitive Indo-European –ehztos and was the suffix used to form adjectives from nouns indicating the possession of a thing or a quality).  Related forms include concatenator & concatenation (nouns), concatenated & concatenating (verbs & adjectives) and concatenative (adjective).  Those who use the undo function on their spreadsheet after concatenating are using the verb deconcatenate and the adjective unconcatenating.  Concatenate the adjective has a longer history than the verb. The adjective first appeared in English in the fifteenth century, the not until the seventeenth.  Catenate, a verb in its own right meaning "to link in a series" also has origins in the 1800s.  Concatenate is a verb & adjective, concatenated, concatenating are verbs and concatenation is a noun; the noun plural is concatenations.

Lotus 123/G running under OS/2 1.2, 1989.

Concatenate is the favorite big word of most accountants, the others preferring avoidance.  For most people not engaged in certain specialised fields, it’s only when using a spreadsheet that the chance exists to use the word concatenate although it’s now often optional, Microsoft in Excel 2016 having added the CONCAT function which does all that CONCATENATE ever did.  The old command remains as a courtesy to those (1) who think the old ways are best or (2) have a stash of macros and add-ins laden with the text but there’s no guarantee both will continue to co-exist in future versions.  Both IBM and Microsoft have often had short and long versions of commands in software.  From the earliest versions of PC-DOS and MS-DOS, there were pairs like copy/cpy and delete/del which behaved identically.

The spreadsheet is regarded as the original “killer app”; the software which suddenly made rational the purchase of a computer for those not before seduced or at least convinced.  The first spreadsheet which really was a viable piece of horizontal-market shrink-wrap was Visicalc which, like the hardware on which it ran now seems limited but, unlike the operating system on which it ran, is conceptually identical and visually, vaguely similar to the latest releases.  Visicalc, launched in 1979 on the Apple II, two years before the IBM PC went on sale, came first but it was the more ambitious Lotus 1-2-3 which gained critical mass, assuming almost from its 1983 debut a market dominance which would last more than a decade.  By 1989, the standard office environment for those running PCs was overwhelmingly the Lotus 123 2.x / WordPerfect 5.x combination, the nerdiest operations perhaps adding the dreaded dBASE III Plus.

Microsoft Windows 3.0, 1990.

In what was one of the early disruptions in the business, things quickly changed.  In 1990, Microsoft Windows 3.0 was introduced, an unstable operating environment bolted on to DOS and soon famous for its UAEs (Unrecoverable Applications Errors), the BSODs (blue screen of death) of the era.  Fragile it may have been but it made the PC usable for real people in a way a command-line based user interface like DOS never did and by the time Windows 3.1 arrived in 1992, the move was on.  Microsoft were ready and Windows 3.1, combined with the updated Excel and Word for Windows sounded the death knell for Lotus 1-2-3 and WordPerfect, both of which were murdered, dBASE more of a suicide as any user of dBASE IV will attest.  The old programs would struggle on, under new ownership, for years, Lotus 1-2-3 lasting until the twenty-first century and a much diminished WordPerfect to this day though neither would ever regain their place in the commercial mainstream.

A concatenation of images of variations in Lindsay Lohan's hair color.

Both failed adequately to react to Windows 3.0, WordPerfect pursuing an evolutionary development of their text-based platform while Lotus followed what turned out to be the right technology but the wrong company.  Almost from the start, Lotus had been besieged by user requests for a way to allow spreadsheets to be bigger and that needed a way for the program to access more memory.  Because of (1) the way DOS was written and (2) the memory address limitations of the early (80x86 & 80x88) hardware, not even all of the 1 MB nominally available could be used and it took not long for spreadsheet users to exhaust what was.  New hardware (80286 & 80386) made more memory available but DOS, really a brutish file-loader, couldn’t see it and the costs of re-equipping with more capable hardware and software combinations were, in the 1980s, high.  There were quick and dirty fixes.  One was a cooperative venture between Lotus, Intel & Microsoft which published an expanded memory specification (LIM EMS), a clever trick allowing access to 4 MB of memory but which brought problems of its own.  Most users continued to create multiple sheets, linking them in a variety of ways, a complexity which was often error prone and, as things grew, increasingly difficult to debug.  It wasn’t just megalomaniacs who longed for everything in one big sheet.

IBM OS/2 2.1, 1993.

Windows 3.0 may not have impressed Lotus but OS/2, Microsoft’s slated long-term replacement for both DOS and Windows certainly did.  Available already with 16 MB of memory, later versions of OS/2 promised 4 GB, a big number then and enough even in 2021 for what most people do with spreadsheets, most of the time.  Lotus nailed 1-2-3’s colors to the OS/2 mast, the first version for the new platform, 123/G (for graphical), released in 1989 and running only on OS/2, did what it claimed and users were soon delighted by the sight (if not the speed) of the spread of their giant sheets.  Unfortunately, users were few because buyers of OS/2 were scarce, their reluctance not helped by Microsoft’s sudden change of operating system direction.  As surprised as everybody else at the massive success of Windows 3.0 and 3.1, Microsoft announced that instead of continuing their co-development of OS/2 with IBM, they were proceeding with Windows as a stand-alone product; existing versions of OS/2 on sale and under development (versions 1 & 2) would be handed back to IBM to pursue while Microsoft would work on their next release which was to have been called OS/2 3.0.  This was the product which would in 1993 be released as Windows NT 3.1. 

It was a high-risk strategy.  In the early 1990s, IBM was years away from its near-death experiences and was the industry behemoth; having them as a partner was not without difficulties but to make an enemy of them was riskier still.  The potential reward however was compelling.  The revenue stream from Windows would flow wholly to Microsoft and, more conspiratorially, having exclusive control of the operating system and its secrets meant the possibility to tweak its own software offerings so they would run better than the competition.  There is of course no suggestion Microsoft ever did that.  All depended on (1) Windows continuing its sales success and (2) the newer versions maintaining the cost/performance advantage over OS/2 which would prevent IBM’s product gaining critical mass.  That is exactly what happened.

Microsoft Windows NT 4.0, 1997.

While OS/2 technically was good and the compatibility issues feared by many never existed to the extent claimed, it simply didn’t offer enough of an advantage over Windows 3.x to justify what would for many be a significant cost in hardware, software and training.  Nor, as the track record with thing like the PCjr demonstrated, were IBM very good at selling stuff unless it was in lots of thousands to big corporations.  Microsoft offered things users were actually interested in, like free fonts whereas IBM fiddled around with exotica like installable file systems (IFS), a concept remote from the lives of most.  Compared with the actually clunky looking Windows 3.x, OS/2 with its IFS, pre-emptive multi-tasking and object-oriented user interface looked like the future of computing and so it was but Windows NT (ex OS/2 3.0) turned out to be a better path.  By the time Windows 95 was released in 1995, Microsoft had won the consumer war and within two years, Windows NT had laid the foundation not only to dominate the desktop in the twenty-first century but to displace Novell and others in the lucrative server market which underpinned the rapidly growing parts of the market, networks (WANs and LANs) and the internet.  In this clash of titans, WordPerfect, dBASE and Lotus were collateral damage.