An LWN status update

In order to grow, far more people need to know about LWN. LWN needs to get featured on Slashdot regularly. Right now, that's impossible, because of the lockout period. By the time LWN news is readable to nonsubscribers, it is generally obsolete. It is admittedly better coverage, but being old news, ultimately, not that useful, since those people have already read the same news elsewhere. Similarly, I've often wanted to forward LWN links to people. I cannot, since those people do not have accounts. I know at least one of the people to whom I would forward links would subscribe (indeed, my girlfriend subscribed after reading LWN over my shoulder).

I strongly believe that in order to grow, LWN needs to eat its own dog food, and open up a little bit. I do not know the best way to do this. At the very least, subscribers should have an "e-mail a link" feature, which lets one specific article be readable to nonsubscribers. There should also be a related way to submit an article to Slashdot, or other blogs, with nonsubscribers being able to view that article. On the other end, you could keep the main pages locked (front/security/...), while keeping the actual articles open. In this way, only subscribers could find the articles in any easy way, but a forwarded or linked article would be automatically readable. The system would easily be cheatable (someone could set up www.links-to-lwn.net), but ultimately, I do not believe that most of the readerbase would cheat, although the most outragous instances of cheating could be caught, especially if the linked URLs had the subscriber's username embedded in them in some form.

A second issue is that I'm not convinced that the path of growth is the correct one. I essentially read LWN for Corbet's coverage, and I wouldn't mind seeing him draw a nice salary, and have an appropriate support staff (bookkeeping, etc.). The remaining editors are less critical to the core business. The outside editors vary, although it is not at all obvious that they are at all motivated by money, so it is not clear if raising payments would help (LWN is considered a fairly respectable place to publish). Some work needs to be done on the LWN codebase, and while it is not clear if the community would get involved in helping the LWN coding effort, there is not harm in releasing the code (unless I missed something, this has not happened yet). A model of some community involvement, in either the non-critical articles, or in the codebase, may allivate some of the work. If the staff could be cut from four down to two, the salaries would increase from underpayed to pretty decent. Similarly, if you extend the scope of LWN, you'll either need more staff (and proportionally more income), or to overwork the current staff even more. Neither seems like a good solution.

I also agree that link forwarding would help. I have at least one friend who I think would subscribe. Convincing him to subscribe would be much easier if I could forward him cool links occasionally, rather than trying to tell him verbally about LWN.

So I think you've got the right sort of content -- well written, topical, informative -- to be quite successful. The issue then comes down to selling/marketing that to people so they at least know they can subscribe, and the benefit of doing so.

I'm ambivilent about the idea of a two-week lock on articles; I think being able to point people at articles is one of the best forms of advertising, and it's often rather less relevant two weeks later. Maybe the "a subscriber sent you access to this one article" approach might be a viable compromise.

It is possibly worth considering whether LWN subscriptions can be sold in a more traditional subscription fashion. Perhaps with an emailed copy sent out (PDF; or maybe a URL) to subscribers who want one, when the weekly edition is published. And/or trying to "partner" with various Linux distributors to sell LWN subscriptions (I'm thinking the likes of EverythingLinux.com.au at least as much as I'm thinking of RedHat/Novell/etc).

Anyway I continue to hope you find a way to make it work in a sustainable fashion.

Ewen

PS: Fixing the cookie issue would be nice too. An updated cookie if, eg, the cookie has less than a week left, and subscriber has more time left than that, would be fine.

By opening to far up, you will remove all pressure to subscribe, except for the "feel good" donation aspect. In principle, this is a sustainable model, but it has been explored by LWN in the past without overwhelming success.

By closing up more, you will increase the value of subscription, at the same time as you make recruiting new readers harder (since the open content is the primary advertising channel - sorry Google). Closing up to much also feels orthogonal to spirit of the communites and subject matter that LWN covers.

The current model, with a one week lock-up period for viewing articles for non-subscribers, is trying to strike a balance between the two extremes. There are a couple of problems with the current approach:

• It is hard for subscribers to send links to non-subscribers. This hampers network effects that would be very nice to have: It would make the content more useful for the subscribers (they can share it earlier), and it would work as word-of-mouth-advertising for LWN.
• The regular lock-up period makes it easy for non-subscribers to just stay one week behind the subscribers. (Analysis of the web server logs could probably tell how many readers who are in this category.)

• Add a feature that enables subscibers to send special "opened links" to friends and colleagues. The resulting page would say "This page is recommended to you by subscriber XXX" at the top, but would otherwise be similar to a regular article. This would create some possibilites for abuse, but not greater than sharing accounts pose today.
• Add some random jitter to the lock-up periods so that some articles will be locked-up for two weeks, and some only by, say, three days. This would mainly make things more difficult for "regular one-week-behind-readers", not for random visitors. My guess is that it would create a little more insentive for some non-subscribers to subscribe. If you combine this with withholding the information on when the content will be freely available, you increase the pain to the addicts, but, sadly, you would also inconvenience the random visitors a little bit.

--
Dag Asheim
Linpro - the leading Linux and OSS provider in Norway

So it's not a product problem, it's related to the way you made this product accessible. Some ideas below, lots of IMHO's suppressed for clarity :)

1-week lockout it's not the best approach, as a lot of people can wait seven days, whilst you cannot reach a wider audience by giving away some on-the-news articles; you should evaluate if increasing the lockout period to 2-4 weeks and made some hot article immediately available could attract more casual user at first, and eventually made them into paid users.

There are some excellent series of article on LWN, like the Grumpy Editor Guide, cannot you revamp them (if needed) and use to attract more visitors? You could possibly end slashdotted :)

Also, I imagine that in the corporate world you can go to your boss and ask for a 200 $/year subscription more easily than for a 54 $/year one, because "if it's so cheap cannot be so good". Such corporate-subscriber could also receive a PDF version of the newsletter via e-mail, so they can show their boss there's something real in exchange.

Last point: in Latin there's a statement like "nomen omen" which could be translated in "a name defines everything" (yes, ancient Latin was one of the most compacted languages ever): Lwn is a lot more than a newsletter, and from the opening page this should be clear. Keep the name, add a slogan :)

It also makes people feel it's ok to only pay $5 per month, after all, they are professional hackers (except for all y'all who aren't, but I'll come back to that), and that's the rate LWN.net deems appropriate, so of course it's ok. Also, "project leader" seems reserved for people who are helped by LWN.net in their position of responsibility and should feel obligated to pay a little more, as opposed to simply being able to afford it.

If you're not a hacker (and not a project leader, manager or similar where LWN.net helps you in your capacity), you'd probably still pick the "professional hacker" option, since you don't feel like calling yourself a "project leader" is appropriate.

As heretic as this may sound to some ("I don't wanna lose my LWN professional hacker status!"), I suggest switching to neutral descriptions, or perhaps no descriptions at all except for the $2.50 one. That you could call "starving linux enthusiast" or some such, and then simply have unnamed "you decide" subscription rates (e.g. $5, $10, $15, $20), maybe tied to some incentive to choose higher rather than lower.

- Micropayments: Maybe instead paying by "days of subscription". You could provide a link to someone else and your subscription ends one day earlier.

- Donating days/weeks/months to authors of articles we like. The thought of prizes is an interesting twist. Maybe a Hall of Fame :)

- Allowing people to subscribe for more money seems like a good plan.

- It can't hurt being Slashdotted every now and them. A link that only works for people coming from Slashdot would be interesting. Or perhaps open some articles to people registered with other sites, to increase networking effects.

- Subscriber-only Tshirts! :)

Keep up the good work.

To increase the base of readers (and subscribers), it may be interesting to translate some articles in other languages like French, Spanish, Italian, ... These articles will be of course available for subscribers only. The problem is : Who will do the translation ?
It could be done by the community (any volunteer ?) or we could ask professionals to do it. But how much does cost such translations ?

Just my 2 (euro)cents :-)))

Bernard.

So why don't you start a 'Design a LWN t-shirt' competition, and see if someone can come up with a nicer logo for both the site and t-shirts? Who knows, you might be pleasantly surprised.

Also, what is it with hackers and black shirts? If you go to thinkgeek.com or any other site that sells geek shirts, you are hard pressed to find any white shirts. Come on, not all geeks are into heavy metal :-) Some of us live in countries where the sun shines hard, and black is hardly photoreflective...

What if we could (when subscribing, or maybe at any time) put extra money into a pot to get specific articles written? So maybe I login, and there's a list made by LWN editors of stuff they could do but weren't necessarily planning for any time very soon. e.g. "Grumpy Editor on VOIP" and I can pay for one or more $10 "votes" to influence what gets written. LWN sets a threshold (I'm guessing this would be in the ballpark of $1000) and when it triggers, those votes vanish and the article gets written. If your $10 vote sits around lonely in an unwanted item (e.g. you turn out to the be one of only three people who wants an editorial on "Why Emacs is better than vi") you could be permitted to move it to something else.

Then I'd guess I expect the article to be locked to subscribers, plus voters if you decided to allow non-subscriber voters, and you'd maybe email all the voters to tell them "You got your wish! Check out the article here"

Anyway, this seems as though it could expand LWN and importantly it would make the expansion proportional to funding, because I know a lot of companies have great plans that go askew only because they guessed wrongly about expansion (either thinking too small and drowning or too big and spending too much too soon).

Oh, and I suppose companies might be interested in putting a lot more than $10 to get something highlighted. I don't know if LWN's principles would be bent too far by Foo Linux Inc. giving them $5000 to write an article about something that Foo Linux happens to be really good at...

Could you give some info on the demographics? Are we all single, 30something geeks or what?

What percentage of your subscriber base is international?

Methinks you need a small marketing campaign. What would it take to grow your subscriber base to 30,000 or even 300,000?

Thoughts:

There must be at least 750 User Groups worldwide?
http://lugww.counter.li.org/ lists about 680.

Give them 20 free subs each. There, your potential userbase is now 15,000. Let's say only 20% renew, you have now doubled your paid subscription level to 6000 in one year.

Encourage them to sign up others. Give a 20% discount when they do so.

Strike a deal, if all renew, give them another 20 free subs.
Voila. you are in the 30,000 ballpark.

Give them a stake: ask them to volunteer articles in lieu of fees. This decreases your costs and increases your subs.

This also increases LWN's international relevance.

Have fun!

To improve things, I think you need to make LWN more visible:

* Take some existing articles, and see if oreillynet.com would be interested in running them, crediting them to LWN.
* Prepare some marketing material aimed at getting corporate subscriptions. Perhaps a publication such as CIO Magazine would be interested in interviewing you or writing about LWN.
* Could some of your editorials be submitted to cnet or zdnet? Or even the op-ed page of a newspaper?
* Add a higher subscription level aimed at corporate customers.
* Submit Slashdot items about longer LWN articles.
* The suggestion of trading a day of subscription to be able to e-mail a sponsored link is a good one; I'd certainly take advantage of it.

1) T-Shirts are always a good idea :-). I like the earlier suggestions to have a contest to redesign the logo. I would certainly pay for it, I am not expecting something for free.

2) More Grump Editor articles. I can hoestly say that the way these are produced makes them invaluable. They cut through all the cruft most other sites have we doing a "review" and get to the real meat. The option of having these (or any other article) downloadable as a PDF (viewable in Evince of course)

3) A series similar to the Grumpy Editor that essentially is a collection of Howto's. For example, a real world small office Linux server howto. Yes, I know there are howto documents, man pages, etc. but someone walking through setting up Samba step by step for the perspective of a technically competent user who may not be too familiar with Linux would be invaluable. I would pay an additional premium if need be for such content, if done in the same style and of the same caliber of the current Grumpy Editor series. I am no newbie (I actually have my Red Hat Certified Engineer Certification) but I also like good groundings in the basics, am more importantly, examples of how it solves real world problems.

Again, my two cents, YMMV

Second, you can add stuff I care about (not just me but others like me, and therefore get some additional readers). Specifically linux clusters. ClusterWorld magazine died this past year, as they could not get sufficient readership to sustain a print magazine. It is not hard to get a cluster/grid column here, and it is quite pertinent to the whole of linux. I don't know if anyone here has noticed, but Linux has achieved world domination in high performance computing. I have written articles for CW as well as my own blogs and others, and it wouldn't be that hard to recruit a few writers for bits like this. Do the same thing for databases, and web servers/portals/...

Just some random thoughts on this.

Here's just a few quick reactions so far, posted here rather than in a bunch of little followups:

• Lots of suggestions of getting slashdotted. It is true that slashdotting happens far less frequency now that the subscription gate is in place. If we extend the subscription period, we probably will put out an occasional article immediately to encourage that sort of thing.

• Subscription offers for user groups. That's been on our list for a while, and tried once or twice at a very low level. We'll probably do something here.

• Some way for subscribers to pass out "read this article free" tokens. That's a new one, we'll have to think on how it would be implemented. Not sure how useful it would be without the ability to post to a mailing list, though.

• Subscription levels: I suppose we could always implement a higher level, though I already feel bad that we don't really have much to offer the project leaders now. Somebody asked about distribution: approximately 18% starving hacker, 76% professional hacker, 7% project leader. Early on, the starving hacker and project leader levels had almost equal numbers of subscribers, but that has changed over time. We'll ponder on whether the levels should be renamed, but I do want the lowest level to remain available; LWN's readership is as global as it gets, and we want to be accessible in places where $5/month is a lot of money.

• Translations to other languages. It's an idea which has been explored, but translations are a lot of work. We get an occasional request for permission to translate an occasional article; we've never said no to that.

• More grumpy editor articles. We want more content like that, but those articles are a lot of work. They also keep well, so a grumpy editor article is usually a sign that said editor is traveling somewhere (this week being an exception to that rule). We'll crank out as many of them as we can.

Thanks for the feedback - we still reading.

So please ask all the subscribers some simple questions like how much you would pay for LWN a year and things like what type of content etc. just to make it interesting. That way you should not feel bad about increasing the rate.

Once you have determined what you think your base likes, try to do the same to the non-subscribers. Perhaps by meeting some of their needs you can convert them.

In addition to mnay of the other ideas, I expect to see a PayPal donation link so if I really like the article I could donate some money to see more of it (either an indepth article or a future article on that topic)

1. Partner with the Free Sofware Foundation: Provide a one-year LWN subscription as an introductory FSF member benefit.

2. Invite Tim O'Reilly out to Colorado for a weekend.

3. Is it possible to only maintain what you are doing, or is some sort of expansion necessary? For example, how about addition of something like the content of LinuxDevices.com? (I understand they are in the Ziff-Davis world.) I really value LWN's thoughtful, well-qualified, no-hidden-agenda, human-voice perspective, and subscriber only revenue supports that. How do you provide more of that while maintaining integrity?

Personally I would like to be able to pay $20.00 a month as that would be still less than one organization wanted from me. I think that the lower end should go up at least by inflation.

If its possible.. I would stagger the free versus pay articles. I wouldnt mind paying for someone to go over the various programmer blogs and point out which articles are important to me. These are usually time sensitive and locking them for a week means that paying for it is important. The weekly items are very important and could go for 2 weeks.

To get a larger audience.. you need to advertise via what people read mostly these days.. blogs. People who pay for this need a vanity item that they can show off.

I haven't gone through the entire list of suggestions, but I thought I'd put these ideas out, feel free to drop any.

Mainly, while the underlying community is all about sharing, as it should be, I don't personally think that analysis of said community or its work should necessarily be free. For me, LWN has been a source of high-level balanced reporting on very important topics, and I would have no objection to see it become paying-only. This may make some uneasy, but it can be a selling angle.

With regards to making some of the analysis available, nothing precludes LWN from allowing companies to sponsor the freeing of content in exchange for more agressive banners in the "freed" content.

Generally, the web site still looks and feels like a community site. While I appreciate the authors' participation in said community, a more polished/ flashy presentation would further highlight the content's value.

Karim

You can't just have the perfect idea and implement it -- you have to pick your top two or three ideas, try them against the current version, implement the ones that work, come up with two or three more ideas, and so on.

Even hotornot.com had to test and change their site around before they got it working well.

You could have just for fun raffles for low cost trivial things, but put some kind of fun community connection to it.

You could also have hardware and components and larger items that the grumpy editor likes. Maybe the hardware vendor would provide the product if asked.

It always seems difficult to know what hardware has open drivers and works best for the open software community, Personally, that is the major consideration when I go hardware hunting. If the grumpy editor provided some information and had a product raffle, well, it might be a fun thing to try. For instance, if grumpy editor found a motherboard that was a good Floss board, subscribers might be willing to enter a raffle to get the product so they could build a box for someone they know.

Grumpy editor might raffle RAM and other components.

I keep thinking of details the would need to be thought about, but I suppose you will understand the concept.

Related to this, maybe you could publish a book, similar to Stallman's "Free Software, Free Society", with a compilation of great LWN articles. It would be cool to trace the evolution of the kernel development process for example.

Have you considered advertising on slashdot (or other places too)? I don't know if it would be worth it, but if you had a text ad with that week's article headlines, I bet you'd draw a lot of people to the site. Seeing the next week's headlines on LWN when I couldn't read them yet was a powerful motivation to subscribe.

Two quick thoughs on "advertising":

1) Maybe you can pull more eyeballs to the site with more authorative content (like the Penguin Gallery, book reviews, and Distributions List). Maybe an additional section could be Free desktop software sorted by category or Grumpy Editor reviewed software (star rated, subscribe to read the full reviews). You may even be able to utilize subscribers to create some of these sections, maybe WiKi style, and leave them free to others as a source of advertising for the rest of the site.

2) I don't think most people object to ads so much as poorly written ads or obtrusive ads (I believe this is why google's ads work so well, they keep editorial control of the format (maybe they would do even better if they took editorial control of the content also, rejecting ads with anoying text) and also why so many people tune into the superbowl....) Maybe you could come up with some sort of directory of software and service providers that you would have editorial control of, but would be provided by sponsors as really clever ads. (e.g. Looking for a HA cluster solution, check with Linux companies a, b, and c)

You might use both of these factors to your advantage by extending the one-week policy by providing immediate teasers of subscriber-only content. For example, look at the lead article in the currently weekly (Finance Managers, part 2). As a teaser, the article title, the intro, a section title, and the section intro show that LWN's original content is valuable and worth a subscription. (A non-clickable table of contents would be helpful too.)

If your value is high-quality content then you can expose the value by giving people a sample up front when it is perceived as fresh and worth a look. When the value has been shown, I think those who appreciate it will subscribe.

Part of each subscription (say $5) compulsorily goes into a pot for that subscriber. All the URLs a subscriber visits have their ID encoded in. Subscribers can send those same URLs to anyone they like; every time it's viewed by someone who isn't them, the page says "Page recommended by person X; please subscribe if you like our stuff" and the pot decreases by one cent. Once the pot is empty, all the URLs stop working. People can donate to recharge their pots if they like.

So, subscribers can share LWN content with their friends legally and immediately, thereby publicising the site, but not in an unlimited way.

Advantages:

- Gets LWN read by non-subscribers in a limited way.
- Gives people an incentive to send LWN links around; if they don't, that pot is "wasted", as it gets emptied and refilled at renewal time.
- Prevents the "Slashdot a free article link" problem - the pot of the person whose link it was would empty very quickly and then the link would stop working.

Gerv

Then some time later our geek leaves the group and thus loses access to LWN. They're highly likely to see the value in an individual subscription -- and may be in a position to convince their next group to get a subscription too :-)

If you do have a marketing push, I think pushing for subscriptions from groups and corporations that are well aware that their members benefit (or profit!) from free software would be much more effective than pushing for individual subscriptions.

So I'd start by putting a standard block of 4 good ads down the right hand
side of the page for non-subscribers.

For your own ads - blow your own trumpet a bit in these LWN updates.
Give some stats about how your ads do and invite more subscribers to place
relevant ads directly with LWN. I place some ads on LWN and am very happy with the cost per view and click thru rate.

Then....

I'd think extending your not-open period to something like 4 weeks
would be fine - but only if you also implement a mechanism for an
easy 1 month free trial subscription. Non-subscribers should be
able to see the first paragraph or two and then a little form that
says:

email: __________________________
[X] Read more now with 1 free months subscription
[ ] Send me a reminder when this article becomes visible in 4 weeks.
[[Submit]]

Those are good and easy ideas.
Unfortunately for you, I recentry extended my subscription period another 12 months.

I would rather there were higher prices that were associated with some kind of return, even something that is obviously unequal to the extra price. Two ideas: (1) when someone posts a comment, make it clear to readers at what level the commenter is subscribed at; or (2) eliminate all variant pricing. Charge everyone $10 or $15, but offer up the option to "sponsor a starving hacker". Then publish the size of the sponsorship pool, and allow anyone who wishes to partly fund their subscription from the pool. If you think too many people will take advantage of this, require them to submit a paragraph about why they think they're deserving, and send those paragraphs to the sponsors (matched up randomly). Or not. You don't want to be forced to spend time moderating them. Perhaps just send the login id of the starving hacker to the sponsor, with no implied obligation on the part of either.

This still feels like squeezing more out of an existing base, though, so increasing the size of your readership is a better plan if feasible.

I would also echo the call to open up your codebase, just to make it possible for people to make suggestions in the form of patches.

I would buy a t-shirt, but you would have to ship it to me, and I would probably only be interested if you offered them when I wanted to resubscribe - now a year away again.

I think ideas like forwarding a link are good. You may get the word of mouth gains. Equally, asking LUGs if they would agree to contact their members on your behalf is interesting. Perhaps some LUG members can suggest if there is an acceptable exchange, such as hosting contact details or even just web addresses for interested LUGs.

From my point of view, you have to become visible to more people. 3k people is nothing compared to the linux user base. I also subscribe to a paper magazine. I do not imagine that this would function with so few subscribers, and wonder if there is a cross advertising deal which you could do. I tend to read news on-line, but prefer the magazine for articles on how to do stuff, like deal with encrypted data systems or set up a WLAN. I think there is little enough overlap that a deal might work. Maybe even some of the Grumpy Editor articles could make it to print, citing LWN.

Or else just take out a page advert in a linux (or computing) publication, giving free access for a month with a code from that magazine. You could even put "this is a free trial", or "your free trial expires in XX days" notices on the page.

I hope you suceed in keeping LWN alive. Please be sure to share your thoughts going forward in the next issue.

1. Put Google adverts in the unused area to the left. Make this a configuration option if you are a financial member.
2. For a fee, by debitting their account, members should be able to d/l an article as a nicely formatted PDF file, or have it emailed to them, or anybody else, as an attachment. Non-members sould be able to do the same, but for a higher fee.
3. Collate the major articles, such as the "Grumpy Editor's" series, every month and produce an e-magazine in both screen and print formats distributed as pdf files. This could carry some adverts and earn extra subscription revenue. I'd subcribe to this in a trice because much of the LWN content is head and shoulders above many print magazines which cost approximately $100 or more per year.
4. Request firmly that you be given quid pro quo links back to LWN from the e-journals to which you link. Similarly the home pages of the Linux distributions to which you link in the Distributions listing linked in the top RH corner of the LWN header.
5. Set up a "share float" starting with the existing subscribers to finance the exercise.
6. Spend some money advertising in Linux oriented print journals.

Also, it should be easier to donate. At the time you were deciding whether to do subscriptions you could only donate by credit card, so I didn't, and then you went to subscriptions which I didn't want to do. Is it still that way? I should be able to either pay online by check or just mail a check.

I like the idea of being able to donate to specific "funds". This could include the fund to get a certain article written as suggested earlier. There could be a "yacht fund" for people who think the editors deserve more money without expecting it to be spent on improving the site. How about this: Have a "free-as-in-beer" fund which determines how long the free version is delayed. Evaluate it every three months, and if there was less than, say, $500 donated that quarter, there is no free version at all; if less than $1000, it's a two-week delay (for everyone); $2000 buys a ten-day delay, and so on. I think that would appeal to anarchist types like me who would like to get some benefit to the common good out of a donation rather than helping only the subscribers and the private-property paradigm. It would be interesting way to test the effectiveness of the subscription-based system and a donation-based system at the same time.

I think before considering a longer lockout period you should reflect very carefully on what you are trying to achieve. If your targets are increasing revenue and improving financial sustainability (as opposed to the perhaps more emotionally satisfying goal of punishing us freeloaders) I seriously doubt longer lockouts are going to help and they might well hurt LWN's chances of long term survival.

Most people who haven't subscribed by now probably don't find LWN valuable enough or don't possess suitable credit cards (in my case it's the former, I like reading LWN's kernel articles now and then and, while they're great, I don't feel they justify the hassle of subscribing). How many of these people can you force to subscribe by reducing the value of the free edition? I suspect it would be quite a small percentage.

Non-subscribers at least know of LWN's existence and like it enough to visit occasionally. That makes us, I think, the best hope of selling more subscriptions in the future. We might change our minds about subscribing, get more Linux-related jobs or, at least, spread word about LWN to others who might become readers.

Now, my suggestion to help finance LWN is giving the possibility of reading the free edition without any lockout by riddling it with ads as much as you like. That would solve the problem of not being able to send links to the content and would increase readership, hopefully without hurting subscriptions.