Went and watched the Mother of All Demos after reading your post.
Holy. Crap.
Holy crap.
My mind is blown.
Yet again I’ve been made to realize I know nothing.
-FrL-
I’ll give them points for funding research - though they are about the only company with money to do it these days. But it’s too early to tell if any of this research results in innovation that we can use, on the order of the transistor or the laser. (Or the mouse, ethernet and GUI from Xerox Parc.) This isn’t criticism, just that it doesn’t count as innovation yet. (I saw tons of stuff from Bell Labs research just as cool that never went anywhere, so I’m a bit cynical.)
It looks like that came about at about the same time as Visual Basic. It’s difficult to determine who was working on it first. Of course it can be argued that Microsoft didn’t innovate VB, because it was based on a concept called Project Ruby that they bought from someone else.
I did a bit of discussion about this with a friend. Some things that were brought up:
Microsoft invented the scroll wheel on mice and also the tilt wheel.
LINQ in .NET 3.0 is undoubtable an innovation but it’s not out yet so I’m leery of what the end result is.
We were surprised by how genuinely hard it was to think of any innovations that have made it into commercial products.
On a more personal side, some people working in fields similar to mine at Microsoft have come up with some amazingly creating stuff. Check out Ken Hinckley’s tablet whacking, “Foreground and Background Interaction with Sensor-enhanced Mobile Devices” and Scriboli. Andy Wilson has touchlight, play anywhere and X wand. Mary Czerwinski has clipping lists, paper lens and scalable fabric. Patrick Baudish has High density cursor (which I was always stunned took so long to be invented) and drag and pop. My friend also mentions that MSR is huge in geometry compression and musical comprehension which is what he’s working on at the moment.
Who, in your view, is doing research of a higher quality, and why?
Google and Yahoo are doing lots of research as well. All three companies have been hiring some of the biggest names in computer science lately. Google doesn’t publish as much as the other two, but it’s not clear to me that any one is better than the others. Google and Yahoo seem to have been better at turning research into money than Microsoft has, though that could change in the future. Microsoft is likely doing more research, simply because they have their fingers in more pots (so to speak) than Google and Yahoo. Even in this, Microsoft is not innovating; rather, they are following Google’s lead.
What is innovation, anyway? If a group of researchers at Yahoo come up with a new algorithm for displaying ads that increases Yahoo’s revenue by 5%, is that innovation? I would say yes, but that’s not innovation that the consumers will ever know or care about. It’s definitely a usable product of research, though, and it’s something that happens not infrequently.
Buh? Microsoft Research was established in 1991, 7 years before Google was privately incorporated. How can you say that Google “innovated” the idea of doing research that isn’t directly product focused? I’m not saying that MS did either (Bell Labs, p’raps?), but c’mon; to say that Microsoft is “following Google’s lead” is just silly.
By the sort of standards that are getting applied to Microsoft here, Google aren’t innovative either because Altavista made a search engine first.
It’s tough tracking down cites for other companies, but Microsoft launched its Certified Professional program in 1992 - I wonder if any other software company realized the value in expanding the brand by having technicians certified to work with their products before then.
(No doubt hardware companies such as IBM had trained techs to work on their systems, but even then I wonder if they offered certification, or just trained internal workers.)
So far the best cite I can find was that Novell changed the name of their CNE in 1993, so it’s certainly possible that it pre-dates the MCP.
Dunno. I just figure claiming “Microsoft is innovative because they spend umpteen-billions on R&D” is a spurious relationship, at best.
I assume he means that a lot of Microsoft’s current “innovations” are following Googles lead - not that Google invented research. The ads at the bottom of this screen are an example of Google’s innovation - MS is now trying to do the same kind of advertising.
As for non-product focused research, in the good old days all big companies did that - Bell Labs, of course, but also IBM, GTE and Xerox that I know of off the top of my head. Remember, the transistor was invented before Mr. Bill was even born.
HyperCard, 1987.
Thanks for that; the Wikipedia article on it is very interesting. As a whole, I’m not sure if it is directly comparable to VB in many ways, but it certainly embodies a number of the same concepts and methods of working, therefore I think we can say that VB was not significantly innovative.
Although I haven’t watched the entire video, it looks like Doug Englebart demonstrated that concept in 1968, Mother of all Demos, previously cited in this thread.
Now that is what I had in mind. It is a very useful, innovative feature that others copied from MS, not the other way around.
To hijack my own thread a bit, I’d like to know if there were any innovations in .NET that made it back to J2EE? I guess that you could say that multiple language support, i.e. several languages compiling down to the same byte code is a feature that seems to be moving into the Java world. As I understand it, thinks like jRuby and Jython and Groovy are all languages that get translated into JVM byte code. I may be wrong about that, so feel free to correct me.
Rob
Java 1.5 adds a lot of features that were previously seen in C#: generic types, typesafe enums, automatic boxing/unboxing, a for-each loop, and annotations (a.k.a. attributes in .NET). I don’t know if .NET really “innovated” any of those, but attributes and boxing weren’t in widespread use before its release.
People have wanted generics in Java since the 1.0 release. The idea first came in C++ AFAIK. The idea of an enumerated type goes back to Pascal, if not further. Given Java’s emphasis on type safety, it’s not a surprise that Java’s implementation would be type-safe(besides, for many years there was a pattern called the “Typesafe enumeration pattern” that was used in Java. 1.5 just cleaned up the syntax, really).
The for-each loop has shown up in several scripting languages before, hasn’t it? As to autoboxing and annotations, I have no idea where they originated.
foreach is definitely old, perl has had it for ages.
IBM’s System/38 in 1978 as well as it’s follow on AS400 in 1988 both are multi-language compilation into a virtual instruction set. The machines, at their core are a virtual machine environment (which is why they have gone through so many CPU changes over the years w/o needing to recompile programs).
It’s going to be pretty difficult to find much of anything that is truly innovative from any company because so much is incremental improvement.
To bring this back to a more general discussion, there is nothing wrong with companies not innovating. Very few PC companies innovate any more, except in sales channels. Monopolies for the most part suck at innovation, since they have so much to protect. And I say that as someone who used to work for Ma Bell. Monopolies feel like they should be innovating, so they fund labs which spin off project which never get used in the company (like Xerox Parc) or they market someone else’s work as innovation. When I was in grad school, IBM was laughed at for a marketing campaign that claimed they invented virtual memory. Microsoft is really no different. Good lab, but do you think anything that comes out of there that might disrupt their virtual monopoly will ever get adopted?
Polymorphic types have been common in functional languages for donkeys years.