For the past 20+ years – all the way back to when it was still privately held — I’ve periodically gotten up to speed on Progress Software. I’m trying again now, and to that end dropped by yesterday for a chat with Jeff Stamen. I’ll give a brief overview now – which is probably all I’m qualified to do right now anyway – and then loop back with more detailed info after I get it.
After a reorganization at the beginning of this (November) fiscal year, the vast majority of Progress’ products fall into one of five buckets, which I shall glibly refer to in decreasing order of size as “Progress Classic,” “SOA,” “drivers,” “memory-centric,” and “EasyAsk.” Here’s a quick overview of each.
Progress Classic is focused on the old DBMS/4GL combo, with the principal brand name being “OpenEdge.” This business is 2/3 VARs, 1/3 enterprises, typically enterprises that were first introduced to the technology via VARs. The 4GL is just historically a great product, if a little outmoded in one or the other way.
The DBMS is noteworthy in that it’s extremely easy to administer. In addition, Jeff claims that even for OLTP there’s commonly a 7:1 ratio in database size vs. Oracle for the same data, due to, for example, variable record lengths. Such factors are not unheard of in data warehousing, of course; but for OLTP his comment was quite the jaw-dropper. On the other hand, OLTP databases aren’t all that big (Progress’s biggest is 2 terabytes), so it’s not clear how great the practical significance of such statistics is.
Incidentally, the ability to scale to that size is one recent enhancement. Another is Java-friendliness; there’s now a JVM, and Progress supports Java stored procedures. Come to think of it, the latter is also mainly a performance/scalability feature in practice, although it might also provide comfort to those users not wanting to be locked into the Progress OpenEdge engine.
Overall growth in this division right now is sloooow. Big VARs continue to see some growth, but smaller ones dropping off aren’t necessarily being replaced.
SOA was put together from a lot of pieces – the inhouse development of Sonic’s stuff, the Actional acquisition, and a bunch of others. I’ll reserve comment until I know more, except to point out that Progress doesn’t seem to have yet connected the dots between its history of application development excellence and the need for really great and simple composite app development tools. Growth here seems solid.
Drivers = DataDirect, which has previously been an independent company, a part of the frequently renamed Sage/Intersolv/Merant, and before that originally an independent company named Pioneer and know for its Q+E product. Basically, DataDirect pioneered ODBC in collaboration with Microsoft, and is still the default supplier of ODBC/JDBC drivers. (Already in the 1990s these were just as performant as native database access, or so close as to make no difference, but to this day many people don’t realize that.) Now the product set has XML mixed in, via Stylus Studio, which is one of the few parts of Progress’s product line that actually seems to be vigorously marketed. Overall, the division seems to be growing nicely, something else that’s a relatively well-kept secret. But then, I’m a charter member of the club that believes DataDirect will maintain its market dominance for as long as John Goodson runs development there.
Memory-centric = what used to be known as Progress Real-Time Division, more or less. Only part of that is doing well. The object-caching and object DBMS products are being deemphasized and are showing small, declining revenue. This is despite some notable product success in the past. Brand names and predecessor companies covered by that include ObjectStore, Object Design, eXcelon (those are all the same thing, basically), and Persistence.
The Apama products, however, seem to be doing nicely in their specialty markets – financial trading, intelligence, telecom. A number of ultra-low-latency technologies seem to be doing well in those areas, as the need-for-speed seems not to be abating.
Note to self: See how Streambase is doing these days.
And then there is EasyAsk. Way back in 1983, I helped Artificial Intelligence Corporation raise scads of money on the strength of its natural language query product Intellect. For a variety of reasons, that never amounted to much, although it was the base for what became a public company called Trinzic. Anyhow, founder Larry Harris tried again, and the result product eventually became EasyAsk and was acquired by Progress. The whole thing is quite incestuous – Larry was on the Progress board, Jeff was an original investor in AI Corp. as well as being in effect a technical cofounder of Progress (despite moving on to other things), etc.
Be that as it may, Progress has some interesting thoughts about the possibilities for EasyAsk.