Lambda the Ultimate

Recently doing research into Wiki's, I've gained a renewed appreciation for the discussion that went on (and still goes on somewhat...) in Ward Cunningham's original Wiki dedicated to pattern discussions. Topics are organized by topic using Wiki words, and there is usually a vibrant discussion that goes on in the page. These discussions are sort of hard to follow since the wiki doesn't have built-in support for discussion threading. On the other hand, discussions are expected to be refactored into the body of the topic's documented by the community. Wikipedia refines this concept a bit by having a talk page behind every document (they are separate but connected).

LtU on the other hand, is more like a newsgroup: you can post a forum topic, but the body of the topic is rarely edited (if only to make corrections) after the topic is started. The discussion is not re-factored into the body of the topic, while some topics are repeatedly revived in slightly different ways and cannot be consolidated into one discussion because refactoring isn't supported. Some topics act kind of like articles, but content is necessarily added as comments since only the owner of the topic can edit the body.

So what's my point? Obviously Ward's wiki is not as popular as it used to be, while LtU is a growing community. Is this because what people actually want are forums and not wiki's? I see a lot of value in the wiki approach myself, and would like my so-called cake and eat it to (that is to say wiki topics with threaded discussions and refactoring).

Anyways just a thought.

The more papers I read, the more I see the term "higher order" applied to seemingly anything. We have HO (high order) programming, HO programming languages, HO functions, HO modules. Those readily come to mind, although I am confident I can find other applications of HO to other terms in computer science.

So what *is* the exact meaning of "higher order." Does it have one precise definition? Or maybe one broader, generally applicable definition? Or is the term simply misused?

Thanks much.

Scott

It's currently not possible to suggest a quotation, as the CGI isn't found.

At least, directing the LtU-blog's RSS-feed (http://lambda-the-ultimate.org/rss.xml) to Twitter, e.g. via a service like TwitterFeed (http://twitterfeed.com/), would be a nice extension.

Best wishes,
fbahr

I am skeptical about this idea getting approved, and even then further skeptical of it actually being used, but I think that LtU could benefit from some extra tags. In particular, Usability/Human Factors would be a nice tag to have, perhaps breaking it into two tags: Human Factors (Real Usability) and Wadler's Law :)

In addition, I think forum old timers are complaining quite a bit lately about stuff being "off-topic". Rather than try to plaster the problem with a Rules&Regulations document, the best way to solve this problem is through some sort of "network of trust". Currently, a network of trust already exists. The old timers trust the moderators to make the right front page content choices. However, some stuff is "maybe off-topic" and others are "definitely off-topic". Moderators could tag things with such, and old-timers can have their newsreaders set-up to filter out definitely off-topic and maybe even maybe off-topics.

When I follow a link to a comment on the tracker page and the comment has spilled to a second page (as on this page), the tracker's comment link takes me to the first page of the story, not the spill page.

While spam is a Bad Thing, it occasionally has the effect of drawing my attention to a page that I've never seen before, bringing old topics to the front for some fresh attention.

The current design of Lambda the Ultimate makes it quite difficult to locate old topics. The use of Search helps if you already have a topic in mind, but if your mind is at all like mine then you often browse LtU hoping only to find the topics at the top of the heap interesting rather than knowing in advance your destination.

I suggest adding support for bringing random topics to the top of that heap.

The possibility my mind seems to be stuck on is adding to the 'Active forum topics' and 'New forum topics' a third column entitled 'Random forum topics' that includes five random topics.

If the idea is accepted, these topics could be chosen randomly on a periodic basis (so everyone sees the same five random topics), and perhaps on a rotating basis (e.g. each of the five topics rotates off the top and a new random topic is injected periodically at the bottom). Alternatively, they could be random on a per-view basis. Some experimentation here is possible.

Are there other thoughts on this possibility?

Hello All,

I have no idea, how should we deal with spam or other inappropriate content.

I miss somehow a button to signal quickly such content. Some other forum engines have such things. I don't know the implementation details.

Regards.

PS. There are two spam topics now, the node 3301 (a klonopin spam) and the node 3300 (a viagra spam).

I think it's high time to have a good, possibly long, and decidedly newbie-friendly meta-discussion about Lambda the Ultimate. Of particular interest is the goals of the site, suitable topics of discussion, and general etiquette.

As I've said before, I have been bothered by the way "off topic" discussions have been handled lately. Moreover, it appears that Ehud's recent front-page post, while accurate, has brought site participation to a virtual standstill.

Again, some of this policing is necessary to maintain the quality of this site, but the civil libertarian in me would like to gracefully accommodate wider participation while maintaining attractive discussions and overall site content.

Let me open this discussion with an extremely astute observation by Philip Wadler:

My experience on the Haskell committee caused me to formulate 'Wadler's Law': time spent in debating a feature doubles as one moves down the following list:

* Semantics
* Syntax
* Lexical syntax
* Lexical syntax of comments

The reason for this is quite simple: it's easy to debate the lexical syntax of comments, whereas it's hard to make meaningful insights into semantics, especially for a language that is as esoteric as Haskell was 20 years ago.

Naturally, more people will be able to offer an informed opinion on the latter topics, even though we shouldn't care as much as we move down the list. (and many of us don't)

Thus, if we model human behavior as a Markov chain, even assuming that each individual person is (somewhat) less likely to add to a lengthy discussion of such trivialities, it's only natural that there will be much more discussion generated on the easy topics. LtU is no exception to this rule.

I absolutely love it when we are graced with the participation of some of LtU's highest-profile members, including Philip Wadler, Lennart Augustsson, Tim Sweeney, Conal Elliott, and Oleg. (who needs no last name) These people are all easily smarter and more knowledgeable than myself, and it makes me very happy that we can attract such participation from time to time.

Of course, there are plenty of other members that fall into the same category, that don't (yet) have quite the same level of fame. Anton van Straaten and neelk immediately come to mind. I don't intend to slight anybody by acknowledging these particular members; my choices reveal more about my own personal (and often obvious) programming biases than anything else.

The elite should also realize that due to Wadler's Law, the more intelligent or abstruse the comment or post, the harder and more time consuming it is to produce a meaningful response. So don't feel too slighted if you don't get as much response as you hoped for, or some noob makes a patently silly or overly longwinded comment. If an appropriate response to such a comment is obvious to me, believe me, I've got your back. Of course, the better I understand your work, the better I will be able to help.

Let me wrap this lengthy post up with one final, but critical point: LtU is not only about the elite, who we welcome with open arms, but also relative newbies like myself, or even total newbies that have no background but have the interest and enthusiasm to stick around and learn something.

If you fall into this last category, your first priority should be to figure out the right questions to ask. Figure out how to provoke us into sharing what we know. And if you think you have some insight, by all means, try.

In the mean time, we'll try to cut you some slack, but try not to be too obnoxious. If you can correct Oleg, by all means, go for it. I definitely want to read it. But beware, the odds are not in your favor. You definitely want to double and triple check your work.

In concurrence with our illustrious and over-worked dictator for life, Ehud Lamm, I would direct everybody to re-read the Spirit Page. Dominic Fox's testimonial is particularly compelling. He deserves to be LtU-famous as well, for very successfully being a total noob.

If you go back and re-read Dominic's comments in the archives, it is unfortunate that he hasn't been around much lately. I suspect that growing demands from his family and his job keep him quite busy these days.

So, in conclusion, I want to hear about your perception of LtU. My goal here is to make LtU better. Any insights into or questions about goals, topics, ettiquette, or even site mechanics are particularly welcome. How can we attract both elite experts and enthusiastic novices without alienating the experts or intimidating the novices?

I think LtU attracts a lot of really bright people. It's thus a shame that detailed discussions regarding design and implementation of programming languages and compilers are considered "out of scope" in the policy document.

Would it be possible to perhaps create a separate forum for these discussions? And/or maybe establish an IRC channel that could be announced on LtU (I'd be happy to set one up)?

Thanks,
-s