Awaiting the Day When Everyone Writes Software
nytimes.com
nytimes.com
Isn't this a bit like saying that architects don't know what kind of houses people want because they spend all day poring over diagrams? Or like saying landscapers don't know what people want because they spend all their time landscaping?
I hate to break it to the New York Times, but programmers are actual people. We're not that different from other people. No really!
It's called COBOL.
(from ca. 1960, that's what people really thought back then)
Heaping layers of abstraction onto each other is NOT something that will inevitably yield "do what I mean" functionality.
What the article describes as Simonyi's "revolutionary idea" is no different than CASE, MDD and MDA - the same (at least) 20 year old pipe dream in various guises that nobody has ever gotten to work except in very narrow fields.
(Full disclosure: I work for Intentional Software. Feel free to ask me about what we do; I'll answer if I can. We're not so much "secretive" anymore as "small and busy.")
When can I try out some of your toys?
How much will that cost. How much will they cost when finished? ("Free" or "Not Free" or "If you have to ask..." are all acceptable answers.)
My interest in what you're doing is largely constrained by my ability to try out your stuff someday.
Would this method be the 'Hungarian Notation'?
"Simonyi introduced the techniques of object-oriented programming that he had learned at Xerox to Microsoft."
"programmers are drowning in ignorance, complexity and error."
...to say Your welcome or F@*k off?
The second snippet is amusing, given the mention of C++ in the first paragraph.
However, I think software has a problem in general with the fact that one shitty system or API can lead to cruft in other pieces that depend on it, leading to generally poor quality. Abstractions and interfaces are designed to prevent some of this, but the worst programmers always find a way to abuse and violate those.
BJARNE STROUSTRUP, the designer of C++, the most
influential programming language of the last 25 years, has
said that “our technological civilization depends on
software.”
I stopped reading there.A concept like this, if it is possible, can work from the other way and help prevent the incompetents from being dangerous. I doubt it will ever prevent people from needing to do low level programming, abstractions leak as the article says. But if done well, it could help people with minimal knowledge create truly useful things. This would help free the truly skilled programmers to work on hard and important problems.
To some extent we are already headed there, although many instances of end-user programming are not considered to be actual programming.
Is someone setting variables in a profile editor programming? How about writing mail-filters? Scripting actions in a larger environment? Using an event recorder to create repeatable macros?
I would say that all of those are in fact programming activities. And I'm fairly certain that there are a large number of people who would disagree.
The thing is that once you look at things in that light, it changes the software you write for the better, it means you provide debugging tools in your application and you look at the softer layers of your application with an eye towards opening them up to end-users.