This prejudice usually indicates the triumph of training over education – you have learned to conceptualize in one, dominating paradigm to the exclusion of all others. Certainly, the same can be said of those who exclusively conceptualize in O-O, or AO or SO or …

1. Polymorphism: depending on how it’s used polymorphism can be the most important mechanism in OOP for supporting change. Not arbitrary change, to be sure, but at least those abstractions conceived of by the original programmer.
2. IDE Refactorings: (not strictly a language mechanism, but the question was posed in the context of PL and IDE support for change). Modern IDEs increasingly support refactorings, both simple and more complex. The support is not always perfect, but it certainly aids the programmer in making changes to the design of software.

Whatever you do with that, the way you protect data or behaviour, it’s up to the language. But from Perl to SmallTalk, data has behaviour and the name of that union is “object”.

While Perl “bless” objects with “module descriptions” and C++ have “classes” to give types to “objects”, both do the same thing to “objects”: it gives them behaviour.

From that, comes the concept of empty objects with only behaviour (helper classes), empty objects with empty behaviour but with style (interfaces), objects without behaviour (PODs), grouping of objects (inheritance, traits, templates). Everything those languages do is with objects, but the way you describe the grouping is via classes or modules, or whatever you want it to be called.

In the end, there’s only one thing that survives: objects and its behaviours. That’s what’s all about OO.

What you complain (and I agree) is how languages implement Object-Oriented design. Stroustroup went wrong when he first designed “C with Classes”, it should have been “Objects in C” to give the right weight to the right things.

But well, that’s how things went because it’s more difficult to do what you say (run view in IDEs) than statically define everything, though it might not be the most logical view for the programmer.

Rest is all useless bashing (I’m sorry for using this term but I couldn’t find a decent word for being pessimist for the sake of it, complaining without considering the same points for the other side of things, …)

For over 10 years, I’ve been reading material – books, magazine articles, blogs, forums – on a hand-held device. JOT has always been a problem; I believe I was “scraping” the site for a while, but this ended up being a bit too much work, and over time, just stopped. I’ve not needed to “scrape” any sites to get material to my reading device for years.

With the changes to JOT announced, I was hoping I could add JOT back to the list of things I can read on my pocket device. I didn’t read this blog post carefully enough though.

Two major issues: PDF only for “research contributions”. and not having a feed for the articles.

re: PDF

PDF is a problem because it doesn’t provide a great reading environment for small devices. Programs like GoodReader for the iPhone provide as good a reading environment as you can possibly get for PDFs, but it can only do so much. A sample JOT article downloaded into GoodReader only displays well in landscape mode, and requires the usual resizing/adjusting/vertical locking that most PDFs do. Frankly, a pain in the ass.

Also, it appears that the articles are only available in PDFs singly, which means I would need to download each one by hand to the device. More pain.

It also seems confusing to have some content in HTML and some content in PDF.

re: feed

A feed is something I asked a previous editor about years ago. A perfect story for me would be to have all of the content published available in the feed. But it appears the “research contributions” aren’t published in the feed, which is avenue of publishing the “non-research contributions”.

I find this distinction between “research” and “non-research” contributions to be odd. Presumably, if I follow you on Twitter and in a feed reader, I’m going to see all the non-research contributions, and will be able to easily read them on a hand-held device. And for all the research contributions, I won’t be notified when they’re available, nor will I be able to read them on my device. Fail.

—

It seems pretty obvious to me that you should publish ALL of your material as HTML, and all contributions should have a feed entry associated with them. If you want to distinguish “research” and “non-research” contributions, do it textually, not via your distribution mechanism. The HTML does not need to be fancy; in fact, it would be nice to have it as simple as possible.

In lieu of getting the articles published in a regular manner, perhaps you could look into just making it easier to consume the “research” contributions on a hand-held device. Which, in the end, is all I really want. This could be as easy as combining all the PDFs into a single PDF file, formatted in such a way as to be readable on a small device – eg, reduce the ENORMOUS whitespace, and use larger fonts.

In the end, sort of feels like the new JOT has stepped fully out of the 1980′s, right into the 1990′s. Very similar to Wired’s recent attempt at producing a version of their magazine for the iPad. Close, but no cigar.