Discussion Locked

Update on community revamp. TRI feedback needed...

0 helpful
45 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

Hey everybody. Remember when we posted those mockups of a promised community revamp a few montsh ago? Well, we understand if you've forgotten. The process has been slower than we hoped, but only ebcause we've been testing and retesting the crap out of the system. So, if you're wondering when the new version of the site is going to launch, well, we have a very loose time frame...

August.

Not very precise, I realize, but there are reasons. Mostly, we have some physical infrastructure projects (read: server moves) that have soaked up all our engineering resources, and we won't get them back until late July, at the earliest. We won't have a real timeline until then. The new hardware should improve things, but while we switch over, there will probably be some hiccups. Fair warning that the summer might be occasionally bumpy as far as site performance goes.

The good news is that our front-end production guy is not part of the infrastructure project, so we have extra time to tweak and refine the presentation layer between now and launch. Beth and Smorty have spent the last two weeks beating the crap out of our beta site, and it's far from over. That's not to say there won't be a few visible seams when we launch--our beta environment can't exactly match or predict the crazy stuff you guys do to our site--but we feel confident the end result will be a dramatic improvement.

So, of course, we're already dreaming of the next phase of enhancements. In my opinion, our site is pretty feature rich--and I would argue TOO feature-rich. I think we've tacked too many bells and whistles on over the years. While all of them were done in an effort to enhance the site, we've got a lumbering giant now with so many hidden capabilities that it can barely move under its own weight. I'd like to strip the site down a great deal, and get at its essence, so more users would be willing to use it. The planned revamp does a fair bit of this, especially in Tech Q&A, but I think we could go farther. On that note, I'd like to solicit some opinions...

1. Does TR need 3 kinds of threads?

Right now, we have blogs, discussions, and Q&A. The redundancy is evident in many places, especially when new users are involved. Should we streamline this, and if so, how?

2. Do we need a granular rating system?

The new Q&A will be a binary rating system. Either it helped, or it didn't, and the negative state is assumed. Articles, downloads, and product reviews have granular (1-5 or 1-10) rating systems. Is that necessary? Should TR go to a simple binary rating scale from everything. Basically, should everything have a "recommend" button, or is the granular rating scale useful?

3. Do we need off-site Linking?

Right now, the TechRepublic Links system lets you save links to URLs outside TechRepublic. Almost nobody uses it for that, however; the bulk of the TR Links directory is taken up by articles and downloads from TR. Is this functionality worth supporting and, if so, why?

4. Do Suggestions need to be personalized?
Right now, the Suggestions in your Workspace are based on tags you've used, articles you read, and whole glut of tracked actions that eat a lot of database space. Think Amazon recommendations. Digg has made it pretty clear that what's popular for a large audience is usually pretty popular with almost every individual in that audience. In other words, strict personalization may be a waste of energy. Could we trade out suggestions for a list of most popular content, or are personalized suggestions worth the effort?

I look forward to the feedback.

45 replies

by jaqui

what about having TRI testing like was said months ago. :(

1) nope, the three threads are a waste of time, if it was all combined it would make it easier to track.

2) yup, the granular system is usefull, specially as applied to articles and downloads.

3) It pretty much duplicates the bookmarks/favorites of people's browsers, right now it's being used to combat the ineffective content search on TR. :p

4) They need to be MORE personalised or done away with.
give the user the ability to EXCLUDE tags / subjects from them, or get rid of the section. I have had frequent times when absolutely none of the suggestions fit my interests at all.

by Jay Garmon

...but that takes engineering resources we don't have available right now. We can't hold up the launch even longer building a test environment. It will happen eventually.

by charliespencer

1. Does TR need 3 kinds of threads? I've been on record for several months that only one type is needed. Kill Q&A and the blogs. Discussions are frequently misused as Q&A, so you might as well drop the formal Q&A. The blogs don't do anything that isn't being done in Discussions, and doing them poorly.

2. Do we need a granular rating system? Maybe a tertiary scale - Good, Bad, Indifferent or something.

3. Do we need off-site Linking? If it adds to the overhead and causes additional workload, drop it.

4. Do Suggestions need to be personalized? Again, if it's causing more work, drop the personalization.

It would be nice if the links in a My Workspace project would NOT also appear in My Workspace Links list. This makes it difficult to determine if you've moved a link to a project; you save it to a project and it's still in the Links list anyway. If a link is going to show up in the Links list, why bother moving it to a Project?

by the admiral

Here are my thinks:

1. Get rid of the doubleclick ads. They are annoying when I get red boxes cause I get doubleclick not found errors.

2. I think three threads can be killed to a single thread with the availability of two or three different topics with different sub topics. Such as Blog, Discussion, Q&A as the Primary, and the subject of the post as the sub. So it says:

Blog: Do we really need all this mess?

3. I have never seen a site that does off site linking that is needed. Most if not all of it never amounts to money in the pocket.

4. Dump the friggin TAGS! They are absolutely annoying and should not be used as a suggestion as for new topics. I often put in:

Winders, Critters and Messes - Oh My!

What can put possibly take out of that other than it is a mess of jumble ripped off and retaped with technojargon from the Wizard of Oz?!

by jdclyde

After the last "please look and comment because the boss is coming tomorrow", that you would have at LEAST had the decency to let us know how that meeting went! X-( [i]how rude...[/i]

There [b]NEEDS[/b] to be a way to choose a discussion vs a TQ&A or the Q&A's will get lost in the mix. If they are not a big draw anymore, get rid of them all together rather than decrease their visibility.

I DID like in the screenshots the ability to at that time decide if it was a discussion or a question.

Blogging. Was it a fad? Do people do it much anymore? I would follow them if a note showed up from my contacts. If I have a contact, (I have a few) I should be notified of ANY activity in discussions/Q&A/Blogs.

I DID like the other day in some history screen it showed that someone added me as a contact. I always thought it would be cool to see WHEN you were added.

Off-site linking. bitbucket.

Personalization, there ARE some things that I will refer to in a few months (like the emoicon list) and without this, it would be impossible to find. I don't CARE about the suggestions, because I often don't have TIME to look them over anyways.

by Jay Garmon

Blog Posts from Your Contacts already exists. It's on the Blogs door. Under the revamp, all PFMC will be merged, and will be sortable by Blogs, Q&A, and Discussions.

Some folks blog religiously, but the numbers are pretty low. I've always been of the opinion that a blog was just a series of posts, so if we listed every thread you ever started, that WAS your blog. There's real debate about that internally, however, as some folks see blogs as a distinct content type with a distinct purpose.

I don't see us resolving the issue any time soon, but I'm interested in what our hardcore users think.

by tig2

I agree with Jaqui- three kinds of threads is unnecessary. Being able to identify if something is a question, blog, or discussion might be handy but I think even that much might be overkill.

I personally do not find the granular scale useful. It is entirely dependant on the engagement level of the person doing the rating. I catch myself listing virtually everything as a 5 on a 1-10 scale. Not because that is appropriate but I often am pressed for time and so don't analyse further. Comes down to "the information is there and I will make use of it." The thing that I would like to see fixed is the point system. I see too many times that the questioner does not feedback on the information received.

I have a file of links that lives on my desktop sorted by type. I use it almost exclusively. The only other hotlink tool I use is on my Google homepage which is always open. I use my Workspace to check contact status, add contacts, and subscribe/unsubscribe threads. I should at some point do a better job with my workspace but haven't gotten that far as yet.

I won't miss the personalisation feature at all. At present, I don't use it. I spend my TR time on what I have determined is a priority and don't bother with much else. The newsletters are a different matter- I often can take a quick read from my email and let that help me determine how I will spend my TR time.

I hope you find this helpful!

by wayne m.

1. Does TR need 3 kinds of threads?
I like consistency. Go with one kind of thread.

2. Do we need a granular rating system?
I would go with a 3-level rating (if, as implied, rating is required): Good, Bad, Indifferent. If rating is not required, just go with 2 levels. I wouldn't bother to rate something as Indifferent if I was not forced to choose. Having 5 - 10 levels of rating is simply beyond my capabilities.

3. Do we need off-site Linking?
I would drop the linking. If someone wants to save a link, he can just save it in his browser and browser and use up his own memory. There is no real need to fill up your servers with that stuff.

4. Do Suggestions need to be personalized?
Right now the suggestions do not seem to give me a very good hit rate, so I don't think personalized suggestions are worth additional effort over generalized suggestions. Maybe expand the "Hot Posts" list and include expanded TR Staff suggestions.

by apotheon

1. Does TR need 3 kinds of threads?

Yes and no. You should combine the handling of the three behind the scenes, but provide different tools for working with each. The notion of just giving a "list by author" interface to discussions has its appeal, and might be appropriate to TR -- that depends on whether TR should actually become a blogging site. The truth of the matter is that for blogging, TR should probably not try to be a blogging site (on top of everything else it's doing), but it shouldn't leave the concept of blogging twisting in the wind, either: I think you would be better served simply by doing the list-by-author thing and providing a way to better integrate external blogs with the community. Having a stand-alone blog functionality just doesn't seem to fit with the TR community as it stands.

I don't agree with Palmetto's bizarre anti-blog fetish, though. I don't know what's up with that. (Hi, Palmetto.)

Q&A shouldn't be seamlessly merged with discussions, or it will cease to be used. Like I said, same back end, different interface. What that actually means in terms of final implementation details is something I'm sure you'll have to work out for yourselves.

One thing that absolutely must be done, though, to improve the overall user experience, is to pare down the categories for discussions and Q&A. I don't think I need thirty different, often overlapping, only barely separate categories under which I can file a discussion thread.

2. Do we need a granular rating system?

Not really. Just provide an increment, a decrement, and a "didn't vote". Have your system total up the values and provide two displayed numbers to people viewing the site: total value and average value (where average ranges between -1 and +1). Anything else is frippery.

If you really want to get fancy, though, you could give people user values that are influenced by the voting they receive on their posts and articles they've had published and so on, then allow their increment and decrement votes to have a weighted effect on the total and average values of stuff on which they've voted. Be very careful with this, however, as a poorly implemented system can backfire. If bad stuff ends up being encouraged over good stuff because of a poorly implemented system, you'll end up with a site that attracts people who are unlikely to find any value in the stuff in your ads, and as a result you could hurt advertising revenue. You're probably better off avoiding the whole weighted value thing.

3. Do we need off-site Linking?

Not really. The linking in general, whether offsite or onsite, is mostly of value to people who don't get out (of TR) much, I think. As such, the off-site linking is probably pretty much entirely superfluous.

4. Do Suggestions need to be personalized?

If you keep them, yes.

Do not -- repeat, [b]do not[/b] -- replace personalized suggestions with "hot discussions" or "popular links" or something stupid like that. It's entirely redundant and unnecessary. If you can't justify personalized suggestions, you shouldn't have anything there at all. If you want a "popular discussions" thingie, just make it a single toned-down link on the first page. People who use it will find it and always know where to get it. People who don't won't care. Cramming a bunch of lemming bait into someone's personal space on TR will just drive people away.

If you keep personalized suggestions, MAKE THEM BETTER. I don't use personalized suggestions, really, but I want to. They're just not very effective. I don't find stuff I want -- or at least not stuff I want that I don't already have -- in my suggestions, so I don't end up using them. I'm sure I'm not the only TR regular in this position. If you could make the algorithm used to determine what's relevant a bit more effective, you'd probably get a lot more use of the personalized suggestions.

by charliespencer

While I do think blogs in general are a waste of electronic resources, I am decidely against the way blogs are implemented on TR.

I can't subscribe to a member's blog the same way I can subscribe to a discussion. Yes, I could use RSS to subscribe to the overall blog but not to an individual posting or the follow-up comments.

Second, the structure of TR Blogs makes those follow-up comments more difficult to track than the "tree" layout in the Discussions area.

Finally, I don't see anything being done in TR Blogs that can't be done in TR Discussions. I freely admit this may be due to a misunderstanding on my part of the Blog format. It looks like both are used for someone to voice an idea, opinion, etc, and for others to respond, and it appears to me the Discussions format does this better than the Blog format. Why waste TR resources and staffers' time on two ways to perform the same function?

by apotheon

"[i]I can't subscribe to a member's blog the same way I can subscribe to a discussion. Yes, I could use RSS to subscribe to the overall blog but not to an individual posting or the follow-up comments.[/i]"
I agree that's a problem, but shooting the blog to put it out of its misery isn't the optimal solution.

"[i]Second, the structure of TR Blogs makes those follow-up comments more difficult to track than the "tree" layout in the Discussions area.[/i]"
Linear discussion is more suited to the purpose of a blog, which is meant to keep focus on the original topic, whereas discussions are more suited to tangents.

"[i]It looks like both are used for someone to voice an idea, opinion, etc, and for others to respond, and it appears to me the Discussions format does this better than the Blog format.[/i]"
Actually, you just described what a blog is supposed to do. A "Discussion" is meant to prompt discussion without that kind of conceptual restriction on the content of responses.

by sleepin'dawg

Start with a statement of mission, keeping it short and simplified to maintain focus. This should clearly define what TR actually is and what purpose it is meant to serve to (a:) management/administrators and (b:) the membership. Somewhere in there, it must be decided which group's needs should be given priority over the general focus. Once that has been clearly defined to everyones satisfaction, then the rest can be dealt with.

1: [b][u]Types of Threads[/u][/b]
When I first approached TR it was with the view of retrieving some long forgotten bits and pieces of technical information that were causing me trouble in terms of current practices. I thought Tech Q&A would answer my needs, not for questions I might want to ask but in grasping a sense of knowledge of current practices. I became very quickly disillusioned. Far too many questions had accumulated and been left in place to offer any viable starting point. Any hopes of acquiring knowledge or information became very much a hit or miss proposition. It took on all the features of an asylum left in the hands of its patients, the administration having long since decamped, in disgust and/or frustration. Witness to that is the long list of unanswered and/or unrated questions still cluttering up server capacity after years, not just months of inactivity. If Tech Q&A is to be maintained it is in serious need of a radical revision and overhaul and administration efforts must be made to clear everything out that does not receive any activity within any given 90 day period.

Blogs: From my viewpoint an indulgence to the egos of those with no confidence in the traits of their personal identities or abilities. Confidence and/or the lack of it, has never been a problem for me and the display of these qualities in others, will only hold my interest in so far as to display the frailties of those, I may care to exploit, for whatever suits my purposes at a later time. A whole section of bandwidth and/or server capacity devoted to this is from my viewpoint, totally unnecessary and superfluous and should be curtailed and terminated, especially any blog that has received no active update or post over a ninety day period. Drivel mostly and who needs it or who cares??? Not me certainly. Begone with it, I say!!!!

Discussions: The salvation of TR and what gives the credence to its existence. I would probably have dropped TR, long ago, if I hadn't stumbled into the discussions and found the knowledge that could be gained from perusing the various forums and the very addictive capacity of Miscellaneous. It is always interesting; more often than not hilarious fun; and often having the similar capacity of watching a train wreck in progress or an endless chain of tailenders on an expressway. A sort of real blood sport restricted to virtual blood.

[b][i][u]VERDICT: [/u][/i][/b]
(a:) Leave discussions as is but clarify that it would be advisable to post any technical questions in any forum other than Miscellaneous until that becomes the final resort.

(b:) Tech Q&A, what I originally came for but is in serious need of revision, a clean out and overhaul. Whether it should be preserved or not is open to question but definitely not in its current capacity.

(c:) Blogs: A useless indulgence to the insecurities of those who are looking to polish their egos and who do not realize that few really care whether they exist or not. The real question might be, does anyone care if they exist beyond the state of their blogs, or not??????

[b][u]2. Do we need a granular rating system?[/b][/u]
I find some sort of rating system useful, whether 1-5 or 1-10 is immaterial however, I often wonder what I miss by passing over some if not most of the lower rated items.

[b][u]3. Do we need off-site Linking?[/u][/b]
Nice but not really necessary.

[b][u]4. Do Suggestions need to be personalized?[/u][/b]
Is it definitely necessary??? I don't know but I can't say I'm a big fan of the seeming disorganization of random Tags. Further more While I like the idea and understand "My Workspace" I can't say I have felt any real need of it and have yet to really use it.

Jay, for what they're worth, my opinions and I can only hope they're helpful towards what you and Shawn hope to achieve. As for Beth, no disrespect intended but I really have no idea what you do but all the indications, received via others, are that you can be as scary as all hell. If I seem to be avoiding you, it's only because I'm such a coward and bleed so easily and so profusely.

[b]Dawg[/b] ]:)

by hal 9000

Firstly my [b]Sincerest Condolences[/b] to the Developer it's bad enough when complete strangers tear your work to pieces but to have your fellow workmates do it must be a bitter pill to sallow.

I see this poor persons job as similar to a [b]Race Teams Mechanics[/b] who spend thousands of hours building a complete race vehicle and then hand it over to the Driver/Rider who does their level best to destroy it and hopefully in the process get to stand on the Podium, but it doesn't happen that often. :_|

Now in answer to your questions

[b]1[/b] No TR certainly doesn't need 3 distinct threads roll them all into one and lets get on with things. As it stands the Discussion threads tend to get used for both discussions and Questions so instead of trying to point new comers to the Q & A section just dump it and place it in a section of the General Discussions, but leave a section in that grouping for Questions so that nay newcomers can find it easily. While you're at it dump the [b]Tech Points[/b] as well as they are about as useful as [b]T_I_T_S on a Bull.[/b]

[b]2[/b] I'm in two minds when it comes to a simple binary system to rate answers particularly when it comes to the Q & A Section maybe 3 options would be better [b]A[/b] It fixed my problem, [b]B[/b] It helped me fix my problem, [b]C[/b] No this was of no use to me. That may be a better idea as it would give others who come along latter some idea of what works and what doesn't. The majority of the problems with the Q & A Section is that the people asking the questions don't know what to ask or what information to give it's very hard to help someone when all you have to work with is [b]My Computer Will not Boot HELP[/b] The answer there could range from a power outage to a dead computer and there is very little to go on.

Example

http://tinyurl.com/r2tue

http://tinyurl.com/rw2qf

Now these 2 questions where asked by the same person on the same day and may or may not apply to the same computer and of course the answers vary depending on what they are applying to.

As for rating White Papers and the like is this Necessary? If the title shows any promise I tend to just grab a copy and look at it latter so I don't get to rate these because I don't read them till much latter generally speaking. If I'm looking for a specific one that's different but the majority of the ones that I grab a copy of are listed in the Newsletters or elsewhere and look interesting from the Title so I save them now and read them latter

[b]3[/b] Do we need Off Site Linking [b]NO[/b] Most people don't use this and it's taking up space on the servers that could be better used elsewhere on the site. The people who use this can place a short Cut on their Desktop or Add to the Favourites or whatever, however on site linking should stay as it's used by the majority of Peers or at least I imagine it would be.

[b]4[/b] Do suggestions need to be personalised? Well at the moment this isn't working overly well and I for one don't use it, others might but if they get the same results that I get when doing a Search I very much doubt that it gets used to any extent so I would tend to say [b]No.[/b]

Col

Message was edited by: admin

by deepsand

1) We definitely need [b]some[/b] sort of mechanism for keeping Q&A identifiably separate from Discussions; picking through the list trying to find the latter among the former is a pain royale.

As for blogs, they do provide a useful function with regards to keeping one's personal writings sequestered and easily locatable. For me, the road block to effectively using such is the inability to organize documents into a meaningful and organized structure. The time linear approach fails to work for me. If I'm overlooking some organizational capabilities please set me straight.

2) No one scale fits all eqaully well; maintain or increase "granularity."

3) While I've thought of using this feature, it seems that so long as I have use of one of "my" machines, with its own bookmarks, I've had no need of it. However, for those who roam frequently and/or widely, this may be quite useful. If it's not being used, then that suggests that it's not needed.

4) Quite frankly, I simply don't have the time to check them all out! So, I'm not sure that my opinion here would be a qualified one.

by deepsand

How is it that so many of those listed in the TR Top 100 show absolutely [b]no activity[/b] as reported on their Profiles?

No Discussions or Q&A participated in; no Blog entries.

No Profile, no History, no Contacts, no Links, nada!

Do these people even exist?

by Jay Garmon

We give extra weighto to blogs, and we allow imported RSS blog feeds. This will go away in the revamp of TR, so these "false positives" should disappear.

by deepsand

According to their Profiles, they have no blogs.

If we can't see or read them, what's the point?

by gadgetgirl

Right, Jay, you asked you're getting it....!! :D

1. 3 kinds of threads? No. TQ &A always seem to end up in discussions, and the points are pointless in the first place. Try staving off the hordes by turning what decent answers are in there into different FAQ sections. Just a thought. Blogs? No, gone off 'em. Had so much trouble trying to get pics to upload, despite destructions from Smorty that I gave up in the end, and abandoned it (through sheer frustration)

2. Considering the number of answers that are un-rated, why bother?

3. Links - nope. Save in the browser.....

4. These have driven me nuts, too. I've tried to tidy up my workspace, but due to the duplications on the links section, I've deleted more than I organised. If it's in the links list, and in your projects, when you delete, it deletes from both.....not very particular, is it?!

Upgrade the search function, and you don't need to personalise the Suggestions in the first place. If we want to read up on something we'll go look for it! Just make sure that the title/tag/content are suitably applicable in the first place, so the search engine can pick it up succinctly.

You asked, you got!

BTW, Jay - about these new emoticons we've been requesting for yonks.......

]:)

GG

by rob mekel

No wonder you gave up :D
If you can't mess up by yourself .... [i]~|~run and hide~|~ what trouble did I get myself now in[/i]

Rob

by hal 9000

Did you forget your [b]Batteries[/b] Again? B-)

Well in an attempt to make you happy tomorrow I'll go looking for that Naked Pic that you asked for and post it if I can find one. :0

Now that will drive Jay nuts waiting for me to post it. :^0 ;)

Col

by rob mekel

Can't do it yourself :D so you needed some help :0 doesn't sound like you.
~|~runs and hide~|~

Rob

[i]edited for damn emoticon typo and remove remark about batteries as Hal 9000 has made that clear. sjeees how many do you need per day GG :0 :D ;) [/i]

by rob mekel

Tnx for the update.

1. Does TR need 3 kinds of threads?
Well discussions goes along every where. So that is one type that is needed. Personaly I think that TQ&A is good to have altho there is a tendency to go into discussion (as Q&A also are in the discussion area) So perhaps it is a good idea to line them up or make them a sub of discussions.

2. Do we need a granular rating system?
A rating system suggests that there is a winning/gaining, or otherwise some honour to gain.
This only stands if there is a consequent rating to it. And that is question of mentality by the users that put up the question.
Personaly I don't need a rating system.

3. Do we need off-site Linking?
I can see the use of it. But do we need a special system for it? My answer would be no. As we all can copy and paste such a link into a posting. Unless you need it for fun as avitars and other stuff. But that's all show and decoration. And more of using other men's sytem as storage for your purpose. Saying this, you can figure out how important TR thinks this linking is as these links use storage on other, not controled by TR, systems it can't be crucial to TR! So it is a need to have thing.

4. Do Suggestions need to be personalized?
Tagging is very handy. What speaks in favour of personal tagging is one of my favourits as it does give me search capability. what brings me to:
[b]Expand search capability[/b] Special in to where has someone put an answer to my posting! As I do like to see what somebody has to say about/to my posting. This in order to answerring a new question/issue to this somebody. [i]I think it is only polite to answer if one take's the effort to answer my postings[/i] And just sending me an email of to what thread has been answerred is not enough.

Rob

[i]Edited for some extra on point no.3[/i]

by sleepin'dawg

They're very big with the sandbox set but are otherwise useless to real people. The "my workspace" seems like a nice idea but I have yet to feel any pressing need for it an I'll bet I'm not in the minority on this. Get back to the KISS princile; if it is used by fewer than 10% consider dumping it; if used by fewer than 5% definitely dump it.

[b]Dawg[/b] ]:)

Message was edited by: admin

by Jay Garmon

If we followed it strictly, we'd dump discussions. Only about 1 in 400 TR members ever actually posts to a discussion. That doesn't mean discussions aren't successful, just that theya ren't as successful as they could be.

by gadgetgirl

A discussion can be classed as successful if it is read. Members don't necessarily have to post to take part - we know that some only post occasionally on their known topic of work.

If it hadn't have been for jd's divorce thread, I'd probably have just been a voyeur myself for the most part - but having found something on which I just couldn't keep my mouth shut, I just jumped right in.

and as you already know, the rest is history......

So, unless you can definitively count the number of times a particular discussion is READ, you really have no idea how successful they ARE!!

GG

by deepsand

Quantitatively speaking what participation level would be considered as being successful?

And, what would be the basis for such?

by sleepin'dawg

Why not tell us your target number and maybe we could all work towards fulfilling it.

[b]Dawg[/b] ]:)

by rob mekel

Did love the new words you came up with. :D

Like this turn in the sub-thread to. :D

Rob

ps
Well I'm off for a long weekend. Have fun.

by charliespencer

"Linear discussion is more suited to the purpose of a blog, which is meant to keep focus on the original topic, whereas discussions are more suited to tangents."

Thanks for the explanation of the difference between discussions and blogs. However, from my tiny perch it looks like TR blogs go off-topic just as often as discussions. That's probably why I perceive they are being used in the same way. Since the author of a TR blog has no editorial control over the comments of others, he has no way to keep it on topic. Do other blog systems have this ability? Maybe this feature is what's needed to differentiate between the two as applied, not as envisioned.

by apotheon

I use a weblog system called WordPress at SOB ( http://sob.apotheon.org ) which, like many such systems, allows a great deal of editorial control. At the very least, such systems tend to provide the ability to delete others' comments -- which is something sorely lacking here.

Some tangential meandering from the focused topic of the original entry is acceptable to the degree that the person authoring the blog decides it is acceptable, of course. A certain linearality of discussion is enforced by the simple fact of the way the conversation is structured, however. Rather than being a tangent, a diversion of topic becomes an evolution of the conversation: the whole conversation pursues a new course, insofar as the author allows it, generally speaking. In TR threaded Discussions, however, thirty different conversations can be spawned that have nothing to do with the original topic of discussion, and that in no way interferes with anyone's ability to chime in on the original topic, or on any of the tangential topics.

I think the lack of separate, useful point to blogs at TR that you're seeing is in large part a result of the TR blogs simply failing to provide a lot of functionality that people expect from their weblog software, and the brokenness of some of the functionality. For instance, TR blogs don't allow for a consistent and reasonably bugless means of using markup -- when you use markup, if you go back to edit it later, it reformats your markup. As another example, there's the lack of overall editing capability. There have historically been some pretty heinous bugs in the RSS import functionality of the blogs, which have since been cleaned up a bit, but even now there doesn't seem to be any way to choose to have an entire blog entry imported unless it's very short: TR blog RSS importing seems to think it's appropriate to import trackback/pingback summaries.

by jdclyde

into the discussion, but still can't buy a shirt.... :(

Oh well, still waiting for the boat pics GG! And now that you know how to post, it will be easy to provide LOTS!

Can't wait for tall ships! I always thought the dozen we get was cool, until I heard about yours! :0 Need a shot of a harbor full of ships!

by Jay Garmon

...would be 2% or rather 1 in 50. While that may not seem like much, that's about the level of return that any extraordinarily successful advertising campaign ever achieves. It's also 8 times our default level. A basically successful ad campaign gets 0.5%, or double our current level.

And, just to be clear, that 1 in 400 figure is uniques against community, which is to say only 1 out of every 400 people who *ever sees a community page on TR* ever posts to a discussion. That discards all the uniques that never venture into discussions/profiles/blogs/Q&As etc. Only about 40% of our site visitors ever enter into a community page.

by deepsand

Depending on the product or service, for some direct campaigns even a 0.5% response rate would be considered outstanding. However, I do not think that such should be considered as a benchmark for gauging the success of discussions.

A direct advertising response requires far less participation on the part of the respondent than does a discussion. And, a response to a direct ad campaign generally leads to a concluding act, whereas a discussion is open-ended.

To truly judge the success of discussions here requires a comparison with the participation rates of those offered by other venues.

by apotheon

Actually, the real metric of success here would be a comparison of the outlay to maintain discussions at TR and the revenue that's generated by them. Of course, determining the effect on revenue can get pretty complex, as there are indirect effects such as how many people end up drawn to non-discussion parts of the site that might keep them coming back and clicking on advertisements, but might have first found the place by a Google search that led to a discussion (for instance). There's a lot of heuristic witchdoctory involved in trying to sort out the "success" for a commercial website with many separate draws for visitors such as TR.

by jdclyde

Because this site is Ad driven, the key is pages VIEWED? I have been seeing it a lot over the last year, and get plenty of peer mails from people that read daily but don't want to post anything themselves for different reasons.

Some are just shy.

Some are new to tech and so they feel like they don't have much to add to the conversations. This is a practice I do at seminars, of hanging on the side of a conversation with the presenters and listen instead of talk. You don't learn if you are the one talking.

Others are worried about posting something that was already posted.

Whatever the reason, they do come and read, and then come back and read more.

THAT should be a goal, to find according to site hits, which topics draw the most hits, and maybe even share that kind of information in TRI.

But the load times going back up sure doesn't help to bring people in. :( Need to get that cleaned up before you try to make a major effort to get MORE people hitting the pages.

Today it is SCREAMING fast, and takes me between 30 to 45 seconds to load this discussion. Recently it has been more than a minute before the page even starts to load. This is from both home or work, using different computers (so it is unlikely something at my end).

Good luck hitting the goals!

by Jay Garmon

My boss's boss is watching this issue like a hawk. The new revamp pages have been CSS streamlined, and our tech production designer has worked hard to ensure that the page won't hang if any one element fails to load (we call in stuff from literally dozens of servers, some of which--like ad servers--we have no control over).

There are a couple of issues, however, that are beyond our power to fix. The first is a basic limitation of every browser--memory failure. Our pages are data-dense, especially long discussion threads. We've seen browsers choke on the data load (surprisingly, Firefox moreso than IE), so that page only half-renders. Not a lot we can do about that.

The second is a mystery to us: Differential load speeds based on locality. We have a lot of anecdotal evidence that users experience page load lags that we don't experience locally, and vice versa. Corresponding with staff that our travelling or based in other cities confirms this. I don't pretend to understand all the inner workings of the Internet, but only about half the time can we trace pageload lags to a slow server. The rest of the time, the data is being slowed somewhere between us and the user. Exactly what's causing that, we're not sure.

by deepsand

You're right, of course; I just did'nt feel like going any deeper than Jay seemed to be looking at the moment. I figured that I'd first address the apples to oranges issue, and get into the rest later, if he seemed so inclined.

As for the complexities here involved, how, for example, might one determine the revenue stream generated by those whose interest was originally generated by something that they saw on a discussion screen, but lead to a deferred action that occurred elsewhere on the site.

by apotheon

Trying to sort out revenue generated by various parts of the site, both directly and indirectly, to cobble together some kind of comprehensive analysis of the dollar value of each part of the site would require tracking the movements of every single person that visited the site, both those with and those without peer/member accounts, even between visits. Once that's accomplished, some kind of processor-expensive data crunching would have to go on with a heuristic analysis of that traffic and the advertising revenue for various pages viewed as compared with points of entry, points of exit, points of return, initial discovery point, and so on.

If they could put together a system that did this effectively and reliably, they could market it and make billions.

edit: . . . and it would surely have to be written in Lisp. They should hire Paul Graham to head up the project if they really want to do that.

by jaqui

quit using the transparent gif images to lay out the discussions, those are just extra images being transferred and rendered.
use spaces or tabs [the html non breaking space for the layout]

second, use text mode advertisements, they require less data transfer and less cpu cycles to render to the window. :p

editing to add:

when it is slowed through the internet, not on either end, it is one of two problems, a ddos / high traffic switch in the routing, or a slow switch in the routing.

yup, there are some switches that have 10 megabit transfer at best and those are often choked by the admin of the location to slow traffic down.

by apotheon

You'd probably be better off using some kind of recursive presentation markup generation to display threaded discussions -- perhaps use list items with a left-border property or something like that to make thread nesting blatantly obvious. This would likely reduce load times as compared with those little GIFs used for the tree display.

On the other hand, each GIF only needs to be loaded once, and many of them are reused, so the performance gain on the client side may be nominal at best. On the other hand, I still think the list item approach (as compared with the GIF approach) would probably reduce server loads, if properly implemented.

by apotheon

I'm pretty sure there's a better way to syntactically represent the semantic structure of Lisp than what's common amongst Lisp dialects, but I haven't seen it (yet). Ruby might give some hints, though it is too object-oriented by intrinsic nature to really serve as too close a guide.

Maybe someday, when I'm a mega-guru like Paul Graham, I'll reinvent the Lisp syntax.

This discussion is closed to new replies.