I suspect Eliezer is targeted specifically due to his tongue-in-cheek presentation of understanding Bayesian statistics as being initiation into a cult. Also due to the author's familiarity with the topic likely specifically as a result of Eliezer's efforts to popularize the subject and his association to him resultantly.
Have used TotalFinder for a couple of years, just tried XtraFinder. Found that the side-by-side view and shortcuts for sliding in the pinnable Finder just didn't work, even after a couple of uninstall/reboot/install/reboot cycles. Has some nice features and seems like it should cover what I want from TotalFinder, but just isn't there yet.
I also don't understand your sentiment for preferring it over TotalFinder... they both feel and act like augmentations. I got the impression that XtraFinder would work alongside the normal finder, but that's not the case.
I know new routers like the D-Link AC5300 DIR-895L/R are forthcoming with faster processors and more streams, but I wanted to take a look at the home wireless router market as it stands today and size up what I think are the best routers for specific uses with an eye towards firmware capabilities and ability to track network usage by-device over time.
Nearly all vendors fail on this, but TP-LINK provides it, ASUS comes close, and ASUS with DD-WRT and YAMon provides it really well for the R7000.
I'm interested to see if there are any devices I'm missing. A nice MIMO 802.11ac device with firmware capability similar to the Pepwave Surf SOHO would be great.
I think this would also lead to a lot of users grabbing a subscription to try out the software who wouldn't have otherwise.
From there of of four things is likely to happen:
1) They decide they don't like it and end their subscription quickly. Since it was cheap, they don't feel cheated and are less likely to smear your brand.
2) They like it and are lazy/preoccupied/like-the-convenience, so they keep the subscription.
3) They like it enough they decide to buy a license so they can save some money long term with upgrades and have more freedom with how they use it.
4) They're not sure if they'll use it enough to justify a continued subscription. If they need it enough, they buy a license without planning to upgrade as a compromise. Maybe they'll decide to upgrade later. If they're not sure, they drop the subscription and maybe buy a month at a time when they need it in the future (you can encourage more frequent users not to do this by offering a 20-25% discount to prepay a year for the service).
Monolithic desktop software like this would pull in more users and revenue with flexible rent/buy options.
I'm also more confident that you're motivated to keep pushing out big meaningful upgrades if you have users that you actually actively have to convince to buy them over what they already have.
No, because users choose to upgrade and own the version they bought outright. You would also have different pricing for new users vs upgraders still, so you'd have the big up front bump from new users.
As you mention, some users don't like the "service" model, so you could present options along these lines (scale values to fit your product):
$120/yr ($10/mo) as a service, or...
$150 to buy now with free minor upgrades and $100 (or scale to be fair for the magnitude of the upgrade) prorated major upgrades.
So I can choose to pay more up front and maybe save a little bit of money in the long run if I choose to keep upgrading, but definitely have access to the software I bought and am familiar with even if I decide I don't use it enough to keep upgrading.
Developers should start prorating upgrade costs instead of setting arbitrary cutoff dates for free upgrades to recent purchasers.
So if you release V1 on Jan 1 2013 and I buy it July 2 2013, I should get the V2 you release Jan 1 2014 for half price. Likewise, someone who buys on Jan 31 2013 should get V2 for 11/12 full price, and someone who buys on Oct 1 2013 should get it for 1/4 full price, etc.
This is in the developer's interest because it discourages people from waiting to buy if they expect a paid upgrade release soon and it leads to fewer requests for free upgrades from people who "just missed" the free upgrade window, for several values of "just missed."
I believe "geek" typically describes someone obsessed with a particular topic and/or with media/games/tech/some-corner-of-popular culture. Conversely, "nerd" typically describes someone obsessed with a particular academic topic and/or with learning in general. It's a significant difference.
Though not typical, it's also worthwhile to remove the "socially awkward" connotations of the use of "nerd" and "geek" terms and ascribe those to "dork" instead. Many self-described geeks and/or nerds don't consider themselves socially handicapped and it's deleterious to society to muddle academic engagement with social impairment.
Great. I gave this some more thought and what I'd actually like, along with a great universal experience, is an email provider that I can really trust. Or rather, an email provider that I don't have to trust. By that I mean that it'd be great if I could get all of my email through Inky and that I'd have the option of having you keep a strongly encrypted copy of those emails while deleting the emails you collect from external providers' servers.
I could have my email everywhere with less worry over wanton snooping over my email archives this way. That would be a very valuable service.
It might require some compromise on data security if I wanted to be able to perform search on my archive via the web service though. Perhaps for people who are willing to give up zero-knowledge you could upsell a service which allows the data to be decrypted server-side in an in-memory store while they're logged in only. I'd be happy without search when using a browser client though.
I really don't want to see ads, but that's preferable over selling my data for a free option. I'd still recommend you give people an ad-free trial (and please make the trial only count days you actually open the app, like Beyond Compare). I'm a free software guy at heart, so the more generous you are with your trial, the more willing I feel to turn a blind eye to opening my wallet for software. Suggestions for how I'd be willing to pay:
1) $20-40/yr for a web service. Most of the world is Windows, so if I'm on the go without my OS X laptop, my phone has died, and I want to check my email I'd have to remember my other email passwords to login to their own services after I got in the habit of using Inky... that wouldn't be good. I'd still want data security, so I'd expect any data that hits your servers to be strongly encrypted in a manner similar to SpiderOak's approach (though I'd prefer Scrypt over PBKDF2). If you open-sourced your client and just sold the service I'd pay more.
2) $50 for the app itself. Less if you want me to pay every year for upgrades.