How do you do the same thing when modifying a local variable? You might try making a copy, modifying that, and then assigning back, but that's boilerplate, and it's easy to get wrong. Worlds are a more primitive mechanism that would be baked into a language, such that building a generic transaction api for local actions on top of worlds would be trivial. The fact that it could look exactly like the transaction api that you currently know and love for remote resources is a feature.
Ok... maybe it's because I like to program the functional way (in imperative languages), but I've never found a reason to rollback local variable changes. That's because local variables don't "fail" and don't need to be rolled back. I see why it's a nice solution but I've never really seen the problem.
What I normally do is: operate on temporaries (or copied parameters), change remote resource (it's safe to just rollback resource changes and exit here in case of problems), then change "the state" when operations cannot fail any more. Is there any real-live problem that needs local variables rollback? (that is shorter / nicer than returning the result from temporary locals) I'd really like to see it.
It's a heavily researched topic right now. The Worlds concept, as far as I can tell, is actually more general than STM and what you described. In STM, there's the global state everyone can see, and then there's local states that try to commit atomically to the global state. With Worlds, I don't think there's one true global state. I think it would give you more explicit control, which I'm not convinced would be a good thing.
The Worlds concept, as far as I can tell, is actually more general than STM and what you described
Where in my comment did I describe Worlds in a limited way? I just gave one small example of one possible application of them, in reply to a comment specifically about their relationship to transactions.
Dumping the current state with deepcopy is probably prohibitively expensive for real world use. Toss in some copy-on-write, and I think you've got something simple and usable...
much more usable than my example, but the copy on write is tricky, which I alluded too. In the case of objects, you need the entire object in order to act upon it, which means you have to copy the whole thing. Of course, if this was an actual language feature, it could be done with some sort of tracing probably, which would make it more efficient. Not sure how you could do that from user code...
I would consider "dammit", "damn it" and "damnit" all to be acceptable. As soon as you leave "damn it" you're in slang territory anyway, so personal variations are acceptable.
(I will also accept "dagnammit", "dognarnit" and "magnabbit".)
"(exclamation) Oldcootism used during great consternation or surprise. Used by 1890’s prospectors, cantankerous old farmers, and young people playing old people on TV in the ‘60’s and ‘70’s."
The period inside the quotation marks isn't clear-cut. Standard American usage is punctuation inside the quotations and is a hold-over from practical considerations of typesetting. Standard British usage is punctuation inside only if part of the quotation itself, outside otherwise. The British style also sees some use in scientific and technical contexts even in the USA, due to being regarded as more logical.
run4yourlives being a programmer as well as, apparently, Canadian, is perfectly entitled to use the more logical British style without regret or hesitation.
Strunk and White say that punctuation goes inside the quotation marks regardless of the context (except for question marks and explanation marks), so if you are listing quotes, then the comma goes on the inside of each quote. On a side note...The serial comma (the last one in a list of items) is most certainly needed (imho?).
3. Use a pair of commas in the middle of a sentence to set off clauses, phrases, and words that are not essential to the meaning of the sentence. Use one comma before to indicate the beginning of the pause and one at the end to indicate the end of the pause.
I have developer sandbox access, but I haven't seen anything about this rollout, nor have I gotten any invites to hand out. If/when I do, I'd be happy to share a few here.
They're established, they can start charging and as far as I know they already are for some categories. They've not got a chicken and egg problem. So I'm not sure what's so 'unbelievably weak'.
On a side note, Joel Spolsky was arguing the same point a couple of weeks back on one of the SO podcasts. He obviously said a lot more, being Joel. I'm paraphrasing poorly but one of his better points was that there's a certain amount of potential money in CL that at the moment is being spent on the very small social good of totally free listings. But he lamented the fact that it used to be spent on the much greater social good of journalism, which is a waste as people don't really need free listings, but they need good journalism for a healthy democracy:
They went on to conclude charging on all categories in CL was inevitable to stop the spammers and perhaps they could use the extra revenue to start funding digital journalism.
That was a great podcast and it had me thinking. But what had me thinking more is when I found out that Joel is also subverting journalism with his million dollar plus in yearly revenue job board: http://jobs.joelonsoftware.com.
That's really a great point but Joel's board is not exactly the same as Craig's list. Joel's point was that Craig removed the revenue from classified ads, when the money was being used for the public good. He acknowledged that the classified ads market was fair game, but he disagrees with Craig's view that he's doing the most good keeping it free.
I think there's a notable difference to Craig's list being free vs Joel's job board costing hundreds of dollars to post. ($400 iirc) With Joel's board he's inviting competitors to under cut him, and there's still a thriving job market outside of his control. Craig's list on the other hand could be argued that by keeping the cost to zero he's making it difficult for anyone to make money off of classified ads because there's always a place to go for free ads. And if nobody else can make money at it then they can't have the option to fund Investigative Journalism and democracy and all that.
I do fail to see why any other industry couldn't fund reporting, but classified ads were a good fit.
Now I really enjoyed the argument he made, but I think Joel should put his money where his mouth is. Joel never argued that Craig should fund reporting as a good business decision, he argued that if Craig wants to serve the greater good (which Craig argues he does) funding reporting is a better way to do it. I'd like to see Joel take a % of his job board and fund the greater good. Part of the problem is how does one help fund investigative reporting? I'd say invest in a newspaper but they might not be the best source. He should run some experiments, hire an investigative blogger, start a grant program for reporter interns, stuff like that. It's his money, but this is obviously a topic he feels strongly about.
And there always will be a place to go for free ads, now that it's been noticed as a good idea for a website; whether it's Craig's implementation that survives or not makes no difference. The cat's out of the bag: classifieds can be free.
A lot of people are making money off of classified ads: it's just that they are individual small-time arbitrage entrepreneurs and small retailers, as well as people getting rid of extra stuff, not newspaper monopolists.
as far as I know they already are for some categories.
The fact that CL is already doing the thing that Godin just descended from the sky to tell us CL should do is, to you, evidence that this isn't 'unbelievably weak'? Will you start reading my blog if I promise to fill it with gems like "What if Google started matching ads with searches?" or "What if Amazon started up an affiliate program?"
For the record, I think Spolsky can be pretty unbelievably lame, as well. People have been talking about how CL should monetize for years.
Even worse is that the slant about being able to use that money to do good in the world (e.g. fund journalism, etc) is almost certainly a swipe from Spolsky's recent craigslist rant on stackoverflow. I guess they might have had the same idea at roughly the same time, but I'm disinclined to think that given Godin's shoddy writing/work lately.
It's almost impossible to charge $1 for something.
As soon as you start charging you need, a secure site, a credit card processor, accountants, lawyers, tax accountants, tax lawyers - repeat for each state/province/country/jurisdiction.
You go from a company with a dozen programmers enjoying what you do to a company of 1000s of employees making sure you have met the requirements for an out of state advertiser in Nowheresville Ak
I have a better idea: what if it costs 1 cent?
That would generate no revenue, but harass posters enough so CraigsList would quickly go out of business.
Have you met Richard Stallman? Not exactly a people person, anyways :)
To be clear, I have nothing but mad respect for Stallman, even though my personal beliefs differ from his. He can be extreme sometimes, but his firm adherence to his beliefs is important. Calling him a traitor is completely out of line.
Yeah, I have, and no, he's definitely not a people person. There's a difference between being a 'people person' and deliberately saying nasty things about other people who have worked their asses off producing free software, though.
You're right in any case: name calling is out of line, for everyone involved, starting with RMS. If he doesn't like Mono or Microsoft or Debian or Apache or whatever, that's fine. But please refrain from personal insults.
I guess my point in saying he's not a people person is to reiterate that folks should just kind of ignore him when he says something like this. Getting personally offended by a Richard Stallman comment is like getting personally offended by a young child wetting the bed. His role is 'tireless proponent of a powerful idea', not 'personable leader'.
After looking at the actual site, the worst thing about it turns out to be the logo. Seriously, the two squids holding tentacles is one of the most disturbing SFW images I've seen in a while.
Anyways, I don't see anything about this that's as horribly unethical as the picture the article is trying to paint. It's certainly not as bad as the Get Satisfaction fiasco, where GS was doing everything short of directly claiming to be an official support channel for companies that had no affiliation. This is just a lame aggregation with a lamer business model. I'm more disappointed in Seth for a lack of creativity, style, and value.
