SDMB Charging for Access - Comments

About This Message Board
121

Poor Algernon might well need more help than we can give if Anthracite catches sight of his gender re-alignment of her.

Kabbes – Yep, but you’re assuming all 500 would cough up and the fee would be $50 – how about if it were 300 at $20 ? Just a very knotty little problem to weigh without more info, IMHO.

I guess that’s why Ed Zoti has gone for the range of prices – no one knows what the response will be. To my mind, the $20 minimum reflects a view from upstairs that they could be looking at 300 paying members.

I don’t think they’d want to overcook the cost of PTP to the users because it undermines the already unpredictable future of the board itself.
Cheesesteak – That’s an interesting hosting option. However, it seems, to me, to be a little premature to suggest any kind of financial commitment based on numbers we can only guess at. Surely, this process of PTP needs to mature over a considerable period until it becomes reasonable to extrapolate an annual income. Then that option could come into consideration.

Given that Dell want a minimum 12 month contract and the experimental nature of all this, at this stage what you suggest might be tricky to get past the Accounts Dep’t.

Just a POV.

122

Damn… as a Newbie I haven’t yet learned the genders of regulars. I’ve tried to make my posts gender neutral but this one slipped. Sorry Anthracite.

123

So much for General Questions.

124

I decided to hold back a few days before offering any thoughts. With my time on the board and number of posts, I’m far from a newbie, but a long way from being one of the hard core.

When I look at my post history, I see that 41% of the boards I’ve participated in were GQ. A whole bunch of the the Cafe and IMHO threads I’ve contributed to have been to add factual information to the discussion at hand. Subtracting a few threads I’ve started looking for specific information, let’s say that 50% of my posting involves answering someone else’s questions.

Would I want to pay for the privilege of dropping in to show off how many facts I know (or can pull up through Google)? Sorry, no. If I’m going to save not-for-profit information media, my local public readio and TV stations get first pick.

But I am of two minds about this. I certainly would’ve kicked in a few bucks had the Board ever actually held a “save the server” drive. And the 25% of posting I do to MPSIMS has SOME value to me.

So if the powers that be want to design some system that I can live with (GQ for free, everything else for pay, or simply asking for voluntary subscriptions) then I’ll probably chip in. Otherwise, it just won’t be worth it.

125

L_C My inclusion of the Dell hosting info was really there only to suggest how much a website such as this should cost to run. Given that info, I can’t imagine how this site could cost more than say $10,000/yr to run, start to finish, including support.

I’d also like to say that if they do charge $20, and only get 300 posting members, that’s still $6,000/yr. With only 300 members, you would need only a tiny server, one that costs much less than $6,000, you could probably get away with one that costs less than $1,000/yr.

That is a dynamic that cannot be forgotten here, fewer members means a smaller server and less bandwidth are needed.

126

How much bandwidth does that Dell hosting solution offer? The IBM managed hosting solution we have costs about that and we are only allocated 1 Mb/s. And that doesn’t include any technical assistance other than a hardware tech we can instruct to do things for us.

Based on my experience, I think it’s reasonable to expect the Boards to cost between $25,000 and $50,000 a year to run, most of that going to the labor required to keep them up.

It would be helpful for the Chicago Reader to provide numbers justifying the charge. It seems quite inescapable to me that the $20 to $50 per year charge is calculated not merely to cover expenses but indeed to make a profit for the Reader. Profits (if they actually materialize, which I doubt) which I rather doubt will be reinvested into the boards.

The Chicago Reader doesn’t have to justify itself to us. And we don’t have to come here. It’s up to them to decide what to do. Trying to support a sagging bottom line by squeezing money out of your fans is a really stupid idea.

In closing, I really hope that if the Reader decides to shut the boards down, they strongly consider transferring them to a non-profit and let the community run them instead of discontinuing them entirely. I’d be much more willing to pay $20 (or even $50) in membership dues to a benevolent society that supports something I care about than give $20 (or even $5) to a company that’s trying to squeeze money out of me for something it used to give me for free.

127

OK: very very rough figures. From the 394 sample members* taken from a reasonable spread of start dates I already have in my database, about 10% have posted more than 0.75 times per day.

Assume 3/4 of these register.

That gives 1350 subscribers out of 18000 current members.

Assume you need $10K per year to cover these.

Then $7.50 would meet your costs and give you a nice $12.50 profit to boot.

I’d say that the population of possible subscribers comes from that 10% of members with 0.75+ posts per day - i.e. about 1800 members. The proportion of these that take up the offer will be inversely related to the cost of the subscription. I think that a 75% take-up of that 10% @ $7.50 is not unreasonable. Reducing it to $5 probably won’t improve matters much - say 80% instead, meaning that you aren’t covering your costs. OTOH increasing it to $10 may reduce your membership below the 55% take-up you need to meet your $10,000 costs.

It’s a tricky trade-off that needs to be made. But you need to play some “silly games with numbers” in order to make any informed guesses at all.

pan

*this is sampled from ALL members registering on a variety of dates from 1999 to 2001, so includes members with 0 posts. It isn’t a bad sample to take conclusions from.

128

** Cheesesteak** – Point taken: You were just providing an example of what can be done commercially at what price. And I agree, it’ interesting especially if the goal of the site remains to fight ignorance rather than to generate profit. I don’t think the intention is to create a surplus but this numbers scenario is little more than a guessing game for everyone – I’d imagine Ed Zoti would be quite sensitive to suggestions of a profit motive, especially given the unpredictable nature of the take-up rate.

** kabbes** - So, rounding that off, you’re saying (maybe) 1,000 members post once a day or above – before taking number crunching further, I’m wondering whether you’ve built your 9 month lifecycle into that ?
KellyM – these are the links Cheesestake provided to the example of Dell:

This is the current configuration:

http://boards.straightdope.com/sdmb/showthread.php?threadid=77679
You seem to understand the Bandwidth game…what are they saying :slight_smile: ?

129

I signed up to pay, and I’m sure the cost will be reasonable.

However, I suspect they’d make more money with either ads or donations.

Doesn’t Fathom run on donations? And they have a lot fewer members, plus better equipment. Did anyone ask OpalCat’s advice?

As for ads, if you got one good account, you’d clear more money than all the subscriptions. When I went to E-Loan, they said they threw money at the web, any place that asked them, and all sites paid off. Every site has people who want loans.

130

No L_C - that’s just a straight unadulterated figure taken from compiling the posts per day of the almost 400 members in my sample.

Those members do spread however from over 2 years old to less than 2 months IIRC. It is, as such, a reasonable cross-section. On those grounds, I feel quite confident in stating that in the wider field of the SD membership, about 10% of the members at any given time have posted at a rate of 0.75 per day or more.

It is these snapshot-members who would be the population most likely to take up a subscription service.

A more complete picture for you:


Posts/day	Total
0-0.25	        325
0.25-0.5	20
0.5-0.75	10
0.75-1	        8
1-1.25	        1
1.25-1.5	5
1.5-1.75	5
1.75-2	        2
2-2.25	        2
2.25-2.5	1
2.5-2.75	2
2.75-3	        1
3-3.25	        1
3.25-3.5	3
3.5-3.75	2
4-4.25	        1
4.25-4.5	1
5-5.25	        1
5.75-6	        1
6.25-6.5	1
10.75-11	1
Grand Total	394

This means that only 8% post at a rate of 1+ posts per day. If this is our potential subscription population, it brings the available subscribers down to under 1500. A 75% take-up rate would then require $10 to make $10K - but would 75% of this population take it up at $10?

And less than 5% post more than twice a day. You can draw your own conclusions.

pan

131

kabbes: Correct me if I’m wrong, but it seems that you are assuming that all or most of the 1+ per day posters would pay up. I doubt that will happen, personally.

Also, I don’t think your equation of $needed / #ofPostersWhoWillPay = Subscription Fee works. We don’t know how many will actually pay, but we do know how much those of us who are willing to pay will pay. Since it seems that there is a consensus that $20 is a reasonable fee, then perhaps we can do it like this: $needed / $20 = #ofPayingPostersNeeded. The questions is: can we clear that number?

If we needed $10,000, then 10,000 / 20 = 500. (Ignoring admin costs, etc.) So, could we get 500 to pay? Obviously, I don’t know how much is really needed.
And in the Wishful Thinking Department: There may be a few committed lurkers out there who will pay up. After all, we are all “lurkers” on PBS and NPR and people donate to them. Just a hope. :slight_smile:

And in the Something I Have Been Wanting to Say, But Haven’t Found the Right Place Department: I suspect that a lot of those high-volume posters who are saying “no way, I wont pay because–choose one: a) it’s the principle of the thing / b) this place will change / c) there are other message boards / d)I can’t afford it / e)other”–I suspect that a lot of those people will pony up when they try a life without the SDMB.

132

Or in other words, kabbes–I am disagreeing with your analysis of the price elasticity of demand. I submit that the demand will be fairly inelastic if it is <= $20.

133

Well, it shouldn’t cost more than a magazine. Can you imagine those with more than one family member on the board? Could be expensive.

Hey, I have a better idea, we could just go on STRIKE & not post/read anymore. Then they would be begging to get us back, maybe even pay us!

134

I think I’m agreeing with what Green Bean just said, but there is a big difference between paying a small amount and paying nothing. Currently, the price is $0.00 per year, and we get a certain number of people registering. Suppose we went to an annual fee of $0.10, that’s right, a dime a year. Now, I’d venture to say that that’s within the means of everyone on the board, even the starving college students. Let’s think of what the registration process looks like now, though: I click the register button, and see that there’s a fee. I’ve got to dig out my credit card, and my checkbook or whatever other system I have for keeping track of my finances. I’m asking myself now how much I trust this website with my credit card number: Will they only take the ten cents, or will they scam me? Later on down the road, I’ve got to remember to go through that again, to pay for my next year. And will the price go up?

Meanwhile, on the Reader’s end, there’s got to be some sort of system set up to process the payments. Even if it’s all automated and no human intercession is needed (except some will be needed even then, when things go wrong), the Reader still has to make an initial investment to put that automated system into place.

All of these considerations will come in to about the same degree, no matter what the price. The difference between people registering at $20 and people registering at $5 is probably a lot less than the difference between $5 and free.

135

No worries, I get mistaken for one or the other on pretty much a regular basis IRL. I was called “sir” twice and “ma’am” thrice at the airport yesterday, so I guess I’m 3/5 female… :slight_smile:

136

If the primary concern about this boards future is related to monetary costs, why not shift it back to the usenet?

Usenet access is easily obtained through google or many other online sources, meaning people who only have web-browsers can still read/post, basic service is probably even bundled with most of your internet services at home. It is with mine.

All the benefits of the board could be realized(except for smilies, but I think we’d cope) in moderated usenet groups. And at little to no cost. A few people may have to bug their ISPs admin to add the groups, but simply charter up a few groups, cite costs of maintaining a web-based forum as the reason for their creation and I’m sure that most admins would carry them. Make them all moderated and have a “subscribe before you can post” rule where someone has to self-register their email address before they can post(cuts down on automated spammers), then let the regular mods do their thing.

alt.straight-dope.general-questions.moderated
alt.straight-dope.great-debates.moderated
alt.straight-dope.bbq-pit.moderated
alt.straight-dope.cafe-society.moderated
etc (the naming scheme doesn’t have to be this, can be anything, these are examples)

I wasn’t around when the transition was made from usenet to these boards, there may have been very good reasons for the web forums over usenet, but if we try again, learning from those experiences, and have moderated groups, do those reasons still hold?

If you turned the machine(s) running the SDMB into NNTP servers, we could all subscribe to those servers as news providers if we didn’t already have news feeds that would carry the groups and the current machines would probably last much longer/be easier to administrate. Being an NNTP server is much easier on a machine than being a webserver. A private news server would give as much control over who/what is posted as we have now. However, I still think moderated groups with a registration method would be sufficient.

If this wasn’t enough of a load reduction the groups could still be registered on the regular usenet so the load would be even more evenly distributed, with many posters/readers picking up feeds from their regular providers. Only those without newsfeeds would have to use the Chicago Reader’s news server.

I have left feedback that I would not pay to post. I don’t post often enough to get value out of it and I already subscribe to several web forums. I’d go back to lurking. I have also left feedback asking why transitioning the boards back to a usenet-based system wasn’t considered an option in Ed’s list.

Steven

137

Dell is offering 21 GB per month in network traffic. That’s not very much: if you average that out over the entire month (30 days of 86400 seconds each), you get an average allowed transfer rate of 8699 bytes/second. A 56k modem is 7168 bytes/second, so what they’re offering you is not much different from a dialup link to the net. (And that’s assuming that the 21GB is outgoing only. If they meter incoming traffic too, you get even less.) That particular hosting solution would almost certainly be inadequate for the Boards for lack of bandwidth.

I can’t find any clear indication on the nature of the Chicago Reader’s current pipe to the Internet. And I can’t think of any good way to gather “page views per day” statistics from the exposed information on the boards. Nonetheless, I shall try.

Examination of a couple of threads I have open here suggests that an average thread view runs between 80 and 120 kilobytes per view. An exceptionally busy thread from the Pit has collected 4900 views in just under three days, or around 700 views a day. Lets say that the average thread gets 200 views a day, and there are (on average) 250 “active” threads daily. (These numbers are WAGs.) So, 250 views of 250 threads of 100 kilobytes each per day works out to 6103 megabytes per day of content served. Over a month that’s 178 gigabytes. The average rate is 74,074 bytes per second. To get that much bandwidth out of Dell would require an additional $340 per month bandwidth charge, and there would likely be frequent additional surcharges for exceeding peak allocated bandwidth even at that level.

(I would invite more informed numbers from those who might have them.)

I assume that the Chicago Reader hosts the Boards currently in their own IT department. The Chicago Reader currently has its own connectivity provided through UUNET. They probably have a T1, and I suspect that the Boards are routinely using about half of that connectivity. (I’m speculating on the size of their pipe; it’s a matter of open record that their ICP is UUNET.) UUNET doesn’t provide pricing information for its hicap services online, and I’m not about to go ask my boss what we pay for our UUNET T1, but it’s a substantial hunk’o’change.

138

Page views you will not find in exposed data, however, I thought the pipe bandwidth was talked about by jdavis a while ago. I cannot find the thread, however, so perhaps not.

You have some pretty good ideas on solutions; I think your ideas should be given some more serious attention.

139

KellyM – If it helps any, I recall mention by Tuba of approaching…one million page views per month. That ain’t no small potatoes! However, I’ve copied (below) part of Anthracite’s post re her experience on her own board which explains how that might work:

Interesting stuff - If I bloody understood it ! :smiley:

140

That would make my WAG off by about a factor of 2. 250 views of 250 threads per day is 1,875,000 thread views per month. So downgrade my previous estimate to 25% of a T1 (a T1 will push 196,608 bytes per second).

I’d call the machine that jdavis described for the boards “woefully underpowered”. And easy enough to do something about: we recently bought a dual P3-866 with a half gig of RAM, 3 36 GB SCSI drives, and a RAID5 controller for about $7000. That should be more than enough power for a bulletin board system. It should be possible to implement a working system for $10,000. That equipment investment should be good for at least two years, so let’s depreciate that over two years: $416/month. A full-bandwidth T1 can be had from t1sales.com for around $795/month including the loop charge through the telco. Software costs for vBulletin are nominal. So, charges exclusive of labor to run the Boards are around $1200 a month, or $14,400 a year.

This is dwarfed by the labor costs to run the boards, which I’d estimate at around 0.8 - 1.2 FTE. Paying someone like that under $30,000 is an insult, so let’s allow $30,000 a year for labor. (This does not include paying admins or mods; just the technical know-how to keep the machine running.) That brings the total to around $45,000 a year.

So, $50,000 a year plus capitalization expenses of around $15,000. Can we do it?