Did anybody think of creating a business out of MythTV? It has some
advantages over existing solutions like TiVo:
* No corporate dogma ('you will have this show recorded and
recommended, because the station paid us to do that')
* Cost
o Cheap extension of diskspace
o No subscription costs, only internet fees
o One-time investment also less
* With DVD drive:
o Built-in DVD player
o Built-in CD "changer" with top quality (cdparanoia, lossless
compression)
* Possible backend server with multiple cheap clients, feeding
several TVs
* No long wires needed
All-deciding for some households: In some cases, there is no
realistic way to get a cable from the satelite dish to the
apartment, and wires within the apartment are a major (very
expensive) problem (hard to get right, and bad solutions are not
accepted). A wireless network may be the only solution to get a
decent feed of TV programs.
* Optionally 2 tuners
* 17" TFT display (smaller case than normal TVs)
Basically, it could be a high-end, middle-cost solution.
MythTV might not be as stable and polished as a mass market expects, but
I guess that won't take long.
The main problem I see is the setup, not only of MythTV, but also of the
hardware. It took me a week to find a card that works under Linux and
making it work. I didn't have success with my Wireless LAN PCI card yet,
although I tried for a week. Maybe I just have bad luck (I don't think I
am *that* clueless), but I see that others also spent a week or more to
get it working. So, that places Linux standard PC solutions basically in
the tinkerer niche, not suitable for average users - think mom and dad,
other people too busy or lazy or incapable to spend time on this.
However, I think they could gain from such solutions, as noted above.
So, my idea is that there should be someone selling PCs optimized for
and pre-installed with MythTV, basically a plug-and-play solution. There
would be several base systems:
* server-only: moderately big case for lots of disks and several tuners
* client-only: small case with good design, fanless, no disk
(network boot), tv out
* TFT client: TFT display with built-in computer and networking
(optionally wireless), client-only
* standalone: as small as possible, as quiet as possible, yet with
enough room for 2 disks and 1-2 tuners.
They would be composed out of selected standard PC components. Slow CPU
(MPEG2 stream comes from hardware) with a mainboard with onboard VGA
incl. TVout, audio and wired ethernet - VIA processors seem appropriate
($100 for proc and such a mainboard). Customers would have the choice of:
* Size and number of disks (40GB - 250GB)
* Type and number of tuners (DVB-S, DVB-C, maybe DVB-T, analog
antenna/cable with hardware encoder)
* Optionally wireless network
* Optionally DVD drive
The hardware would possibly be assembled per customer spec (like cars
and Dell computers) or have common pre-configs. The software would be
installed (esp. incl. hardware drivers) and pre-configured to the
maximum extend (boots into MythTV, the EPG source would be preprogrammed
for the relevant country, network config via DHCP, server discovery via
DNS). The customer basically only has to plug in the cables and run the
autosetup for the tuner. The hassle-freeness is the argument against a
self-built computer solution that we all (current MythTV users) use.
Apart from more flexibility, cost could be an argument in competition to
existing mass-market products. I think that servers and standalone
machines could start at about $600 and clients at about $300, maybe
less. The configurable parts (tuners, disks) should have street prices.
Existing solutions with similar capabilities often have price tags of
$1000 and more.
Design of the cases is very important. No chances with normal computer
cases. A few nice products for the clients, all fanless, but some are
too expensive:
* <http://www.storever.com>
* <http://www.mini-itx.com/store/hush.asp>
* <http://www.mini-itx.com/store/default.asp?c=15#p133> (unless I am
overlookign something, complete only 240 Euro (!) as client with
RAM and remote control)
The following has a nice design for the standalone, but no pratically no
PCI slots (need 2) and a fan: <http://www.mini-itx.com/reviews/enote/>
The Internet connection (if needed) could be a configuration problem -
the user would have to set that up, and it would need to be connected to
the existing network/computer, which could be a hassle to set up (esp.
if there is only a single Windows machine connected to the internet) or
be a technical problem (conflicting IP network addresses, if connected
to existing network). Maybe the EPG/teletext data sent by the stations
could be used for the EPG program data? What about CDDB for CD ripping,
then? Automatic software updates?
Anybody wants to make a business out of that? (I assume that this would
also end up contibuting valuable code to MythTV.) It's nothing for me,
but it would be nice to see such a product and have something to point
people at. Just an idea...
advantages over existing solutions like TiVo:
* No corporate dogma ('you will have this show recorded and
recommended, because the station paid us to do that')
* Cost
o Cheap extension of diskspace
o No subscription costs, only internet fees
o One-time investment also less
* With DVD drive:
o Built-in DVD player
o Built-in CD "changer" with top quality (cdparanoia, lossless
compression)
* Possible backend server with multiple cheap clients, feeding
several TVs
* No long wires needed
All-deciding for some households: In some cases, there is no
realistic way to get a cable from the satelite dish to the
apartment, and wires within the apartment are a major (very
expensive) problem (hard to get right, and bad solutions are not
accepted). A wireless network may be the only solution to get a
decent feed of TV programs.
* Optionally 2 tuners
* 17" TFT display (smaller case than normal TVs)
Basically, it could be a high-end, middle-cost solution.
MythTV might not be as stable and polished as a mass market expects, but
I guess that won't take long.
The main problem I see is the setup, not only of MythTV, but also of the
hardware. It took me a week to find a card that works under Linux and
making it work. I didn't have success with my Wireless LAN PCI card yet,
although I tried for a week. Maybe I just have bad luck (I don't think I
am *that* clueless), but I see that others also spent a week or more to
get it working. So, that places Linux standard PC solutions basically in
the tinkerer niche, not suitable for average users - think mom and dad,
other people too busy or lazy or incapable to spend time on this.
However, I think they could gain from such solutions, as noted above.
So, my idea is that there should be someone selling PCs optimized for
and pre-installed with MythTV, basically a plug-and-play solution. There
would be several base systems:
* server-only: moderately big case for lots of disks and several tuners
* client-only: small case with good design, fanless, no disk
(network boot), tv out
* TFT client: TFT display with built-in computer and networking
(optionally wireless), client-only
* standalone: as small as possible, as quiet as possible, yet with
enough room for 2 disks and 1-2 tuners.
They would be composed out of selected standard PC components. Slow CPU
(MPEG2 stream comes from hardware) with a mainboard with onboard VGA
incl. TVout, audio and wired ethernet - VIA processors seem appropriate
($100 for proc and such a mainboard). Customers would have the choice of:
* Size and number of disks (40GB - 250GB)
* Type and number of tuners (DVB-S, DVB-C, maybe DVB-T, analog
antenna/cable with hardware encoder)
* Optionally wireless network
* Optionally DVD drive
The hardware would possibly be assembled per customer spec (like cars
and Dell computers) or have common pre-configs. The software would be
installed (esp. incl. hardware drivers) and pre-configured to the
maximum extend (boots into MythTV, the EPG source would be preprogrammed
for the relevant country, network config via DHCP, server discovery via
DNS). The customer basically only has to plug in the cables and run the
autosetup for the tuner. The hassle-freeness is the argument against a
self-built computer solution that we all (current MythTV users) use.
Apart from more flexibility, cost could be an argument in competition to
existing mass-market products. I think that servers and standalone
machines could start at about $600 and clients at about $300, maybe
less. The configurable parts (tuners, disks) should have street prices.
Existing solutions with similar capabilities often have price tags of
$1000 and more.
Design of the cases is very important. No chances with normal computer
cases. A few nice products for the clients, all fanless, but some are
too expensive:
* <http://www.storever.com>
* <http://www.mini-itx.com/store/hush.asp>
* <http://www.mini-itx.com/store/default.asp?c=15#p133> (unless I am
overlookign something, complete only 240 Euro (!) as client with
RAM and remote control)
The following has a nice design for the standalone, but no pratically no
PCI slots (need 2) and a fan: <http://www.mini-itx.com/reviews/enote/>
The Internet connection (if needed) could be a configuration problem -
the user would have to set that up, and it would need to be connected to
the existing network/computer, which could be a hassle to set up (esp.
if there is only a single Windows machine connected to the internet) or
be a technical problem (conflicting IP network addresses, if connected
to existing network). Maybe the EPG/teletext data sent by the stations
could be used for the EPG program data? What about CDDB for CD ripping,
then? Automatic software updates?
Anybody wants to make a business out of that? (I assume that this would
also end up contibuting valuable code to MythTV.) It's nothing for me,
but it would be nice to see such a product and have something to point
people at. Just an idea...