7 ms·
I have very little perspective here, but this seems like a low price for /. and SourceForge ($20M). The article states that last year alone the two sites broug
by zheng 14y ago
I have very little perspective here, but this seems like a low price for /. and SourceForge ($20M). The article states that last year alone the two sites brought in that much revenue, but I'd assume they weren't bought if the purchaser didn't think there was growth potential.
- jnorthrop 14y agoAnd conversely, they wouldn't have been sold at 1x annual revenue if the seller didn't think that the future prospects were bleak.
- dredmorbius 14y agoAnd/or if the parent wasn't desperate for cash and found few takers. That said, upside doesn't look all that great.
- debacle 14y ago$20M in revenue, but how much in profit? Slashdot and sourceforge can't be cheap to host.
- viseztrance 14y agoPerhaps, but how much revenue did Instagram have?
- rmc 14y agoThe had millions of users using it for photos when Facebook, who had lots of money, was worried.
- Codhisattva 14y agoSpeculation drove up Instagram's "value". As did a likely bidding war with Google.
- dredmorbius 14y agoAnd it's an established business. No zero-point extrapolation here.
- masterponomo 14y agoYou nailed it. GKNT shareholder here. They have struggled for years to monetize the media business, with little success (profit-wise). Cost of revenue is always very high. Add in a lot of turnover in editorial/content/business leadership (basically everything on the media side) and it has been a whole lot of operating losses. ThinkGeek, meanwhile, has had great year-over-year sales increases, and costs are being brought under control. With the sale of media business, 2012 could be GKNT's first profitable year in many a year (or ever--haven't checked back more than 5 years). Or it could be they finally unloaded the media business (they hired a company to shop it around, BTW) and will now try to sell the e-commerce side (say, to Amazon) and close the doors.
- eli 14y agoI heard $5M.
- seiji 14y agoThey are basically internet retirement communities. They have a stable population until the population starts to die off. Except with a normal retirement home, there are always new old people joining. No sane person under 45 really wants to be around the /. "we hate everything" crowd or whatevertheheck sf.net turned into.
- nn2 14y agoNot sure about sourceforge. It lost its lustre to github, but there are still a lot of good project there. freshmeat/freecode is really great. Lots of great software there (I liked the old name better) Browsing the freecode frontpage is far more satisfying than the HN page because it's all about real code that does something and very little hype and BS.
- astrodust 14y agoReading Sourceforge is not unlike reading magazines at the dentist's office while awaiting a root canal. All the magazines are old, dog-eared, and full of things you don't care about and the environment is extremely disconcerting. This latest redesign takes Sourceforge from Slashdot-flavored ugly into domain-squatter horrendous.
- nn2 14y agoTrue for the html layout, but not the content (the projects/code). For a code site layout in the end is very secondary. Besides github is doing its best to compete on the "largest number of useless buttons/toolbars on the webpage" front with sf. I also found that finding something based on keywords works better on sf than on guthub.
- astrodust 14y agoFor code, layout is everything. Consider how excruciating it is to try and view source code on SourceForge compared with GitHub. On the latter you are literally shown a listing of files with the README already open for any given repository. Sourceforge has this absolutely useless "Download" link where you have to grab an archive, extract it, dump it into your text editor, and only then will you find out it's not what you want. It is not an understatement to suggest that Sourceforge is by people with no idea about open source.
- dmethvin 14y agoMany profitable businesses tail off very slowly over years, throwing off excellent revenue in the process. Last year, for example, AOL made $175 million off 3 million customers who still subscribe to its dial-up business.[1] Those businesses can become even more profitable in the near term if your strategy is to cut marketing/operating costs to the bone and "ride them down". It sounds very mercenary, but that can be the right approach when there is no growth potential. [1] http://www.splatf.com/2012/07/aol-3million-chart/ http://www.splatf.com/2012/07/aol-3million-chart/
- cydonian_monk 14y agoMy dad is one of those 3 million dial-up subscribers - but not by choice. He doesn't use dial-up, hasn't had a PC with a modem in it for several years, but every time he calls AOL to cancel they give him six months to a year of free service.
- cstejerean 14y agoWithout a modem, why even take AOL up on the free six months to a year of service? It just means you'll have to call again in 6 months, and so on.
- dredmorbius 14y agoSubscriber numbers themselves can be useful, even if only to count as possible future revenue or marketing.
- cydonian_monk 14y agoPresumeably because calling every six months or so is easier than telling people your new email address.... I honestly don't understand it myself.
- maxerickson 14y agoThe email service is free.
- Codhisattva 14y agoWell it's different to sell a profitable company than a speculative company. Valuation of a start up is based on speculation while an operating and profitable company would go for around 110% of gross annual revenue in cash. Since it used old school valuation we can assume a few things: * GeekNet content traffic is growing very much * There was no other bidder and no 'bidding war' * GeekNet was highly motivated