My own private idaho (self employ)

[Bill Sloman]

On Tuesday, January 28, 2020 at 11:59:38 AM UTC+11, Rick C wrote:

On Monday, January 27, 2020 at 7:42:30 PM UTC-5, Lasse Langwadt Christensen wrote:
tirsdag den 28. januar 2020 kl. 00.53.22 UTC+1 skrev Rick C:
I've wanted to make some items that are small and cheap, like Chinese made and sold sort of cheap. But I can't find a way to sell they way they do. There could be lots of competition, but not until the volume gets high enough. I would love to contract out the whole thing, but my understanding is I would never see much money from it because of the way they work.


look at things like arduinos, arduino shields, MCU boards,
usb logic analysers, etc. I don't see how anyone makes any money on those. The minutes it gets any traction you can buy a clone on ebay with free shipping
for half your BOM

That's why I said "like Chinese". I mean tap into their process. Have them build it and sell it on eBay and Aliexpress and I get my $0.10 commission or however much. But I've yet to figure out how to do that without being cut out.

Actually, it might be practical to do this if the product has some aspect that requires connecting to the Internet to access a server. That would allow control and monitoring as long as the code is secure enough to not be hacked. That can be done in an FPGA...

I need to file that away for the deluxe version of a product I've wanted to build.

You need to start reading up on public key encrytion.

It's covered in chapter 8 of Davies and Price "Security for Computer Networks" ISBN 0 471 90063 X published in 1984. I mainly bought the book because I'd had an interaction with the authors a few years before the book got published, when they were proposing to make the Teletex protpcol explicitly capable of handling it. The Teletex system was supposed to offer a step up from Telex (and did for a few years in Sweden and Germany).

It's quite a neat system. You need a pair 1024 bit long keys these days, and quantum computers threaten even them, but so far it seems to work.

--
Bill Sloman, Sydney

 

[Tom Gardner]

On 28/01/20 00:59, Rick C wrote:

Actually, it might be practical to do this if the product has some aspect
that requires connecting to the Internet to access a server. That would
allow control and monitoring as long as the code is secure enough to not be
hacked. That can be done in an FPGA...

Dongles are disliked for good reasons.

Virtual dongles (including DRM) are hated because
there are too many "you're left dead in the water"
failure mechanisms.

Start by thinking of internet connectivity, then DDOS
attacks on some part of the infrastructure, then on
your servers, and finally consider servers being
switched off or a company folding.

Learn about Microsoft's PlaysForSure (TM) (sic) music
viz: "DRM servers related to PlaysForSure were turned
off August 31, 2008, meaning that any operating system
upgrade or migration rendered all content unplayable."

This week's hoo-haa is about Sonos speakers.

Customers are rapidly becoming sensitised to the
problem.

 

[Rick C]

On Tuesday, January 28, 2020 at 2:54:19 AM UTC-5, Tom Gardner wrote:

On 28/01/20 00:59, Rick C wrote:
Actually, it might be practical to do this if the product has some aspect
that requires connecting to the Internet to access a server. That would
allow control and monitoring as long as the code is secure enough to not be
hacked. That can be done in an FPGA...

Dongles are disliked for good reasons.

Virtual dongles (including DRM) are hated because
there are too many "you're left dead in the water"
failure mechanisms.

I think you exaggerate the issue. The device only has to be authenticated once.

Start by thinking of internet connectivity, then DDOS
attacks on some part of the infrastructure, then on
your servers, and finally consider servers being
switched off or a company folding.

ok, considered. Now what?

Learn about Microsoft's PlaysForSure (TM) (sic) music
viz: "DRM servers related to PlaysForSure were turned
off August 31, 2008, meaning that any operating system
upgrade or migration rendered all content unplayable."

This week's hoo-haa is about Sonos speakers.

Customers are rapidly becoming sensitised to the
problem.

Ok, thanks.

--

Rick C.

-+- Get 1,000 miles of free Supercharging
-+- Tesla referral code - https://ts.la/richard11209

 

[George Herold]

On Monday, January 27, 2020 at 5:41:44 PM UTC-5, Clifford Heath wrote:

On 28/1/20 7:16 am, George Herold wrote:
Hey here is sorta a crazy idea I had emailing with Clifford H.
How about designing/ building (as kits?) analog front and back
ends for the labjack?
https://labjack.com/

It's a bad idea(*) for a startup to be completely dependent on the
existence of any other single product or company. The Labjack looks like
a packaged version of the plentiful AD8706 modules, with some software.
Why wouldn't you just make those too, so you aren't dependent?

Search AliExpress for AD8706 and you'll see modules with parallel
interface for around $40, with USB for $60-ish. That gets you 8 channels
of 16-bit data at reasonable speed, which should be enough for many of
the lab projects you have in mind.

(*) The problem is that as soon as you become successful at some add-on
product, the company can easily duplicate your work and undercut you.
That's almost instant death when you're relying on their installed base.

Clifford Heath

Hi Clifford. Well first building something like the labjack is not my forte.
So better to let someone else do it. Much of the lab/interface/ hardware
and software space is taken up by national instruments and lab-view.
I find their hardware over priced and hate the software. So pushing an alternative seems like it might not only be good for me.. but also for the
physiics lab community.

I guess I wouldn't worry much about labjack coping me... small market.
If I head dwon this road I would try and contact labjack early on...
see if there are other vendors like me re-selling their stuff.

Anyway it is interesting to think about. at my PPoE I would often get
requests from customers to provide some sort of an experiment where students
could learn instrument/ computer interfacing. Going to physics lab
workshops the way people are doing this these days is either with
Arduino's, maybe a few Rasbery Pi's and labveiw.

One huge problem with this idea, is I'd have to become a labjack expert
and at the moment I'm a novice.

George H.

 

[George Herold]

On Monday, January 27, 2020 at 7:42:30 PM UTC-5, Lasse Langwadt Christensen wrote:

tirsdag den 28. januar 2020 kl. 00.53.22 UTC+1 skrev Rick C:
I've wanted to make some items that are small and cheap, like Chinese made and sold sort of cheap. But I can't find a way to sell they way they do. There could be lots of competition, but not until the volume gets high enough. I would love to contract out the whole thing, but my understanding is I would never see much money from it because of the way they work.


look at things like arduinos, arduino shields, MCU boards,
usb logic analysers, etc. I don't see how anyone makes any money on those.. The minutes it gets any traction you can buy a clone on ebay with free shipping
for half your BOM

You've got places like Spark fun and ada-fruit which seem to be
carving out a niche in this space.

George H.

 

[Sjouke Burry]

On 28.01.20 8:54, Tom Gardner wrote:

On 28/01/20 00:59, Rick C wrote:
Actually, it might be practical to do this if the product has some aspect
that requires connecting to the Internet to access a server. That would
allow control and monitoring as long as the code is secure enough to not be
hacked. That can be done in an FPGA...

Dongles are disliked for good reasons.

Virtual dongles (including DRM) are hated because
there are too many "you're left dead in the water"
failure mechanisms.

Start by thinking of internet connectivity, then DDOS
attacks on some part of the infrastructure, then on
your servers, and finally consider servers being
switched off or a company folding.

Learn about Microsoft's PlaysForSure (TM) (sic) music
viz: "DRM servers related to PlaysForSure were turned
off August 31, 2008, meaning that any operating system
upgrade or migration rendered all content unplayable."

This week's hoo-haa is about Sonos speakers.

Customers are rapidly becoming sensitised to the
problem.

Have (some) links about the sonos problem?
A friend has his sonos apps in disarray , one phone to old,
the other phone its app needs a password, but updating that password
at the sonos site fails, (new password not accepted, try again).
So their radio turned into a doorstop.

 

[Tom Gardner]

On 28/01/20 17:39, Sjouke Burry wrote:

On 28.01.20 8:54, Tom Gardner wrote:
On 28/01/20 00:59, Rick C wrote:
Actually, it might be practical to do this if the product has some aspect
that requires connecting to the Internet to access a server.  That would
allow control and monitoring as long as the code is secure enough to not be
hacked.  That can be done in an FPGA...

Dongles are disliked for good reasons.

Virtual dongles (including DRM) are hated because
there are too many "you're left dead in the water"
failure mechanisms.

Start by thinking of internet connectivity, then DDOS
attacks on some part of the infrastructure, then on
your servers, and finally consider servers being
switched off or a company folding.

Learn about Microsoft's PlaysForSure (TM) (sic) music
viz: "DRM servers related to PlaysForSure were turned
off August 31, 2008, meaning that any operating system
upgrade or migration rendered all content unplayable."

This week's hoo-haa is about Sonos speakers.

Customers are rapidly becoming sensitised to the
problem.

Have (some) links about the sonos problem?
A friend  has his sonos apps in disarray , one phone to old,
the other phone its app needs a password, but updating that password
at the sonos site fails, (new password not accepted, try again).
So their radio turned into a doorstop.

Google is your friend...
https://www.bbc.co.uk/news/technology-51206604
https://us.cnn.com/2020/01/24/cnn-underscored/sonos-legacy-speakers-guide/index.html

I should have emphasised "this /week/". Read comp.risks,
and you will see many other examples of this. Google/Nest
springs to mind.

 

[Lasse Langwadt Christense]

tirsdag den 28. januar 2020 kl. 17.04.53 UTC+1 skrev George Herold:

On Monday, January 27, 2020 at 7:42:30 PM UTC-5, Lasse Langwadt Christensen wrote:
tirsdag den 28. januar 2020 kl. 00.53.22 UTC+1 skrev Rick C:
I've wanted to make some items that are small and cheap, like Chinese made and sold sort of cheap. But I can't find a way to sell they way they do. There could be lots of competition, but not until the volume gets high enough. I would love to contract out the whole thing, but my understanding is I would never see much money from it because of the way they work.


look at things like arduinos, arduino shields, MCU boards,
usb logic analysers, etc. I don't see how anyone makes any money on those. The minutes it gets any traction you can buy a clone on ebay with free shipping
for half your BOM

You've got places like Spark fun and ada-fruit which seem to be
carving out a niche in this space.

which is a bit surprising since much of what they make can be had for
a tenth on ebay complete with a link to sparkfun/adafruit for documentation
and software

 

[Phil Hobbs]

On 2020-01-26 15:31, Phil Hobbs wrote:

On 2020-01-26 06:06, Piglet wrote:
On 26/01/2020 00:15, George Herold wrote:
Sure with pressure as the pusher.  The one thing I hate about any
magnetic valve ideas.. is that it will f with any magnetic measurements.

George H.

Good luck George.

Getting movement without magnetic fields makes me wonder again about
Nitinol* "muscle wire" - always seemed a great solution looking for a
problem to solve. But possibly not right for your app, as far as I
know it is bang-bang only with no proportional control, slow in the
hundreds of milliseconds, and being thermally driven probably a
complete mismatch for cryo N2 valves!

piglet

*For those unfamiliar a shape memory alloy wire that switches between
two states and corresponding lengths with temperature. Passing a
current to self heat makes a sort of electrical-mechanical transducer.



Nitinol goes much faster if you use a thin gauge and dunk it in oil--you
can get audio rates out of it.  I have a couple of Blue Sky 'Collimeter'
units, which are shear-plate collimation testers dithered by a bit of
Nitinol wire driving a flexure pivot.

It'll give you nice smooth bidirectional motion if you have some sort of
feedback.  There's a resistance change associated with the phase
transition, but I don't know how repeatable that is.

Cheers

Phil Hobbs

It might be interesting to mount a Nitinol coil spring coaxially with a
steel one to make rotary motion. The spring rate would let you limit
the torque on the valve, and running the current down one spring and
back up the other would cancel out the magnetic field pretty well. The
parlour trick would be the encoder.

Alternatively a model airplane servo would work--you can turn it off
between motions to get rid of the field.

Or you could put a breakaway shaft extension on the valve, so that Mr
Gorilla just has it come away in his hand without anything breaking
permanently. Something in a nice low-density polystyrene, maybe.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[speff]

On Tuesday, 28 January 2020 11:00:41 UTC-5, George Herold wrote:

On Monday, January 27, 2020 at 5:41:44 PM UTC-5, Clifford Heath wrote:
On 28/1/20 7:16 am, George Herold wrote:
Hey here is sorta a crazy idea I had emailing with Clifford H.
How about designing/ building (as kits?) analog front and back
ends for the labjack?
https://labjack.com/

It's a bad idea(*) for a startup to be completely dependent on the
existence of any other single product or company. The Labjack looks like
a packaged version of the plentiful AD8706 modules, with some software.
Why wouldn't you just make those too, so you aren't dependent?

Search AliExpress for AD8706 and you'll see modules with parallel
interface for around $40, with USB for $60-ish. That gets you 8 channels
of 16-bit data at reasonable speed, which should be enough for many of
the lab projects you have in mind.

(*) The problem is that as soon as you become successful at some add-on
product, the company can easily duplicate your work and undercut you.
That's almost instant death when you're relying on their installed base.

Clifford Heath

Hi Clifford. Well first building something like the labjack is not my forte.
So better to let someone else do it. Much of the lab/interface/ hardware
and software space is taken up by national instruments and lab-view.
I find their hardware over priced and hate the software. So pushing an alternative seems like it might not only be good for me.. but also for the
physiics lab community.

I guess I wouldn't worry much about labjack coping me... small market.
If I head dwon this road I would try and contact labjack early on...
see if there are other vendors like me re-selling their stuff.

Anyway it is interesting to think about. at my PPoE I would often get
requests from customers to provide some sort of an experiment where students
could learn instrument/ computer interfacing. Going to physics lab
workshops the way people are doing this these days is either with
Arduino's, maybe a few Rasbery Pi's and labveiw.

One huge problem with this idea, is I'd have to become a labjack expert
and at the moment I'm a novice.

George H.

Hi, George:-

I guess the important thing is to know the customers and what they're
comfortable with- software they get free or already have and know how
to use, and hardware that they think is good for their application. The
younger guys I have been dealing with like MATLAB/Simulink, for example (despite
free alternatives), and I have yet to find anyone who likes Labview much, but some use it. As an aside, I've been playing around with it, because I ended up
with a little NI data acquisition system that runs off a PXI Windows XP card and
a couple NI USB-powered data acquisition modules. Nice for gathering data in
simple experiments.

This is just an impression but I get the impression it's not that uncommon for
university purchases to go either top end (they have specific funding and are
going to spend it on really nice equipment) or bottom scraping (they've
got very little budget), which doesn't leave much room in the middle. That's
not the educational market though, which I'm not that familiar with. I've
seen a few products like Mr. SQUID, for example, but no idea of that market.

Best regards,
Spehro Pefhany

 

[bitrex]

On 1/27/20 7:42 PM, Lasse Langwadt Christensen wrote:

tirsdag den 28. januar 2020 kl. 00.53.22 UTC+1 skrev Rick C:
I've wanted to make some items that are small and cheap, like Chinese made and sold sort of cheap. But I can't find a way to sell they way they do. There could be lots of competition, but not until the volume gets high enough. I would love to contract out the whole thing, but my understanding is I would never see much money from it because of the way they work.


look at things like arduinos, arduino shields, MCU boards,
usb logic analysers, etc. I don't see how anyone makes any money on those. The minutes it gets any traction you can buy a clone on ebay with free shipping
for half your BOM

Free shipping on that stuff generally takes 2-3 weeks I'll gladly pay a
little more to get the product in 48 hours or sometime before Hell
freezes over.

 

[Rick C]

On Tuesday, January 28, 2020 at 11:06:00 PM UTC-5, bitrex wrote:

On 1/27/20 7:42 PM, Lasse Langwadt Christensen wrote:
tirsdag den 28. januar 2020 kl. 00.53.22 UTC+1 skrev Rick C:
I've wanted to make some items that are small and cheap, like Chinese made and sold sort of cheap. But I can't find a way to sell they way they do. There could be lots of competition, but not until the volume gets high enough. I would love to contract out the whole thing, but my understanding is I would never see much money from it because of the way they work.


look at things like arduinos, arduino shields, MCU boards,
usb logic analysers, etc. I don't see how anyone makes any money on those. The minutes it gets any traction you can buy a clone on ebay with free shipping
for half your BOM



Free shipping on that stuff generally takes 2-3 weeks I'll gladly pay a
little more to get the product in 48 hours or sometime before Hell
freezes over.

When I order that sort of stuff I typically don't have a pressing need. Having a couple of weeks to think about it on the back burner is usually a good thing.

--

Rick C.

-++ Get 1,000 miles of free Supercharging
-++ Tesla referral code - https://ts.la/richard11209

 

[Winfield Hill]

Clifford Heath wrote...

a packaged version of the plentiful AD8706 modules...
Search AliExpress for AD8706 and you'll see modules ...

AD8706?


--
Thanks,
- Win

 

[Michael Kellett]

On 26/01/2020 19:25, George Herold wrote:

On Sunday, January 26, 2020 at 4:22:03 AM UTC-5, Tom Gardner wrote:
On 25/01/20 21:22, George Herold wrote:
The other bad thing about selling stuff is you've got to
spend ~1/2 your time on marketing... maybe more early on.

And all the other administrivia Even if someone
else is doing all that guff, and engineer operates
at full throttle for about 30%-40% of their time.
The rest is "unproductive" meetings etc. Hence in
the absence of better information, multiply your
first estimate by 2.5-3.5

I've seen a couple of startups make one mistake;
I suspect you are too smart, but here it is anyway...

The founder(s) are trapped by the glitz of being
"self-employed CEOs", and spend money on unnecessary
/visible/ trappings, e.g. cars, boardroom furniture,
and the like.

They feel great and "energised", but none of that
gets money in the door.

More successfully...

I've also seen successful R&D consultancies that
only buy the minimum of equipment, i.e. equipment
that is guaranteed to be useful on every project.

Any equipment that doesn't fall into that category
is bought on a project-by-project basis and after
the project it is part of the deliverable to the
client.

Good luck; all startups need that.

Huh, no I won't be purposely buy un-needed stuff.
If I was to do this LN2 thing, the first thing I'd need is
a leak detector. (Or an RGA and pumping station.)

I think I'm mostly trying to talk myself out of this idea.
If I could work 1/2 time for someone else and 1/2+ for myself
that might work.
George h.

Some ideas that may be useful:

I quit a family business 20 years ago to work on my own and it's been
good. I'm not rich but doing OK.
No one bosses me about, I go out with the dog during the day when it
suits me and work when it's raining or dark.

I've done 100% work for other people and while I've had zillions of
ideas I don't make my own products. Still thinking about it but doubt
I'll ever do it - I'm good at what I do but a lot less good at all the
stuff you need to be into to sell stuff and produce it.

I design things, write software, make prototypes and one off test gear
and do some product testing. It pays for loads of fun gear and a pretty
good time.

I don't get hung up about other people making money from my ideas -
there's always another idea coming along during the next dog walk !

I do buy test gear - I quote jobs on a fixed price - its up to the
client to decide if likes the price or not - he doesn't need to know it
covers loads of labour or a new scope.

In the end you have to discover what works for you, but it might be good
to try for a couple of freelance type jobs from existing contacts while
they are still fresh. Nothing builds your confidence like sending out an
invoice.

MK

 

On Tue, 28 Jan 2020 08:04:47 -0800 (PST), George Herold
<ggherold@gmail.com> wrote:

On Monday, January 27, 2020 at 7:42:30 PM UTC-5, Lasse Langwadt Christensen wrote:
tirsdag den 28. januar 2020 kl. 00.53.22 UTC+1 skrev Rick C:
I've wanted to make some items that are small and cheap, like Chinese made and sold sort of cheap. But I can't find a way to sell they way they do. There could be lots of competition, but not until the volume gets high enough. I would love to contract out the whole thing, but my understanding is I would never see much money from it because of the way they work.


look at things like arduinos, arduino shields, MCU boards,
usb logic analysers, etc. I don't see how anyone makes any money on those. The minutes it gets any traction you can buy a clone on ebay with free shipping
for half your BOM

You've got places like Spark fun and ada-fruit which seem to be
carving out a niche in this space.

George H.

Maybe do something exotic and physics-related. Cryo sensor adapters?
Mag field sensors? You know that business.



--

John Larkin Highland Technology, Inc

The cork popped merrily, and Lord Peter rose to his feet.
"Bunter", he said, "I give you a toast. The triumph of Instinct over Reason"

 

[Phil Hobbs]

On 2020-01-29 09:58, Michael Kellett wrote:

On 26/01/2020 19:25, George Herold wrote:
On Sunday, January 26, 2020 at 4:22:03 AM UTC-5, Tom Gardner wrote:
On 25/01/20 21:22, George Herold wrote:
The other bad thing about selling stuff is you've got to
spend ~1/2 your time on marketing... maybe more early on.

And all the other administrivia Even if someone
else is doing all that guff, and engineer operates
at full throttle for about 30%-40% of their time.
The rest is "unproductive" meetings etc. Hence in
the absence of better information, multiply your
first estimate by 2.5-3.5

I've seen a couple of startups make one mistake;
I suspect you are too smart, but here it is anyway...

The founder(s) are trapped by the glitz of being
"self-employed CEOs", and spend money on unnecessary
/visible/ trappings, e.g. cars, boardroom furniture,
and the like.

They feel great and "energised", but none of that
gets money in the door.

More successfully...

I've also seen successful R&D consultancies that
only buy the minimum of equipment, i.e. equipment
that is guaranteed to be useful on every project.

Any equipment that doesn't fall into that category
is bought on a project-by-project basis and after
the project it is part of the deliverable to the
client.

Good luck; all startups need that.

Huh, no I won't be purposely buy un-needed stuff.
If I was to do this LN2 thing, the first thing I'd need is
a leak detector.  (Or an RGA and pumping station.)

I think I'm mostly trying to talk myself out of this idea.
If I could work 1/2 time for someone else and 1/2+ for myself
that might work.
George h.

Some ideas that may be useful:

I quit a family business 20 years ago to work on my own and it's been
good. I'm not rich but doing OK.
No one bosses me about, I go out with the dog during the day when it
suits me and work when it's raining or dark.

 I've done 100% work for other people and while I've had zillions of
ideas I don't make my own products. Still thinking about it but doubt
I'll ever do it - I'm good at what I do but a lot less good at all the
stuff you need to be into to sell stuff and produce it.

I design things, write software, make prototypes and one off test gear
and do some product testing. It pays for loads of fun gear and a pretty
good time.

I don't get hung up about other people making money from my ideas -
there's always another idea coming along during the next dog walk !

I do buy test gear - I quote jobs on a fixed price - its up to the
client to decide if likes the price or not - he doesn't need to know it
covers loads of labour or a new scope.

In the end you have to discover what works for you, but it might be good
to try for a couple of freelance type jobs from existing contacts while
they are still fresh. Nothing builds your confidence like sending out an
invoice.

MK

We've recently been moving from a straight NRE basis into know-how
licensing plus NRE for customization. If you have enough recurring
themes in your designs, you can do your own based on stuff you've
learned on other jobs.

Many of our projects, especially for new customers, include a photon
budget and proof-of-concept (POC) demo as the first two milestones.

These days our POC systems usually consist largely of a collection of
our existing boards (often modded) plus fairly small amounts of
hand-wired custom circuitry and fairly simple optics and mechanics as
required.

Here's the POC from the cathodoluminescence system that I posted a few
months back:
<https://electrooptical.net/www/sed/MPPCphotonCounting.mpg>. It goes
from photon counting up to bright room lights with a twist of a knob,
which we think is pretty slick.

We find that when they see our stuff doing what they wanted, better than
they hoped, cheaper than they expected, the licensing conversation
usually goes pretty smoothly. It helps that our royalty expectations
are fairly modest--we'd way rather do lots of deals that put smiles on
everybody's faces than screw the last dime out of only a few.

YMMV, but a bit of recurring revenue is very comforting.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Clifford Heath]

On 29/1/20 11:35 pm, Winfield Hill wrote:

Clifford Heath wrote...

a packaged version of the plentiful AD8706 modules...
Search AliExpress for AD8706 and you'll see modules ...

AD8706?

Sorry, typo. AD7606. There's also the 24-bit ADS1256, and modules for both.

Here's Analog's selection table, lots of nice devices:
<https://www.analog.com/en/parametricsearch/10994>

CH

 

[Michael Terrell]

On Monday, January 27, 2020 at 5:43:56 PM UTC-5, Bill Sloman wrote:

On Tuesday, January 28, 2020 at 4:02:52 AM UTC+11, jla...@highlandsniptechnology.com wrote:
On Mon, 27 Jan 2020 08:47:12 -0800 (PST), George Herold wrote:

On Sunday, January 26, 2020 at 3:50:04 PM UTC-5, Phil Hobbs wrote:
On 2020-01-26 15:42, George Herold wrote:
On Sunday, January 26, 2020 at 11:16:31 AM UTC-5, jla...@highlandsniptechnology.com wrote:
On Sun, 26 Jan 2020 09:21:58 +0000, Tom Gardner
spamjunk@blueyonder.co.uk> wrote:

On 25/01/20 21:22, George Herold wrote:
The other bad thing about selling stuff is you've got to
spend ~1/2 your time on marketing... maybe more early on.

And all the other administrivia Even if someone
else is doing all that guff, and engineer operates
at full throttle for about 30%-40% of their time.
The rest is "unproductive" meetings etc. Hence in
the absence of better information, multiply your
first estimate by 2.5-3.5

I've seen a couple of startups make one mistake;
I suspect you are too smart, but here it is anyway...

The founder(s) are trapped by the glitz of being
"self-employed CEOs", and spend money on unnecessary
/visible/ trappings, e.g. cars, boardroom furniture,
and the like.

Some people think that the appearance of success (hence buying all
that glitzy expensive stuff) is success. It's the opposite.

One guy that I know of got some investors to join up, and first thing
bought a private jet. Next thing, they fired him.

(One common mistake is to give up too much equity early on, often to
less than savory/competent partners.)


They feel great and "energised", but none of that
gets money in the door.

More successfully...

I've also seen successful R&D consultancies that
only buy the minimum of equipment, i.e. equipment
that is guaranteed to be useful on every project.

General-purpose equipment is good to have around. These days, a suite
of good test equipment isn't very expensive.


Any equipment that doesn't fall into that category
is bought on a project-by-project basis and after
the project it is part of the deliverable to the
client.

NO! Keep all the toys! Well, you can provide custom test sets at large
additional cost. Price them separately; they may find that they want
several.
Yeah, a bad thing about losing my job, was losing most of my toys.
(I've got some power supplies, rigol scope and rigol sig. gen.
and as you say parts are pretty cheap these days.)

George H.



Good luck; all startups need that.

One can certainly be a consultant, and sell the IP and designs and
equipment on a one-time basis. That gets a company started, and some
people can do that forever. But that requires constant selling, and is
often feast-famine stressful. Selling products is a better long-term
force multiplier.
Right, you need a few horses in the stable for sales (at least at the
low volume level I'm contemplating.) to be semi-stable.

The other thing I think about is fun little pcb's and a part's bag/ list.
There's lot's of rather mundane circuits you can think about,
(gain, filtering)
fun ones are rev. biased led spads, and fast edge TDR's

Some standard box for little circuits would be nice.
Phil, what do you pay for your 'stomp' boxes.
(two piece dicast )

They're cheap--five or ten bucks depending on size. BTW once you have a
web site up, you can do pretty good SEO by adding your contact info and
a few keywords to your Usenet posts--SED is widely mirrored. (See below.)
Huh thanks for the SEO tip.
Re: SED mirroring. I do find it weird how many times I'm searching some
electronics thing and I get sent to a few years old SED discussion.
Sometimes that's good, cause it's a discussion I missed or skipped.

George S. Herold
unemployed instrument builder

Sometimes I google a subject and find my own old SED posts.

Want to do some occasional research for us? Email me.

Jim Thompson seems to be dead, so John Larkin will have to find some other right-wing lunatic to to tell just who has just e-mailed him privately.

James Arthur might serve.

I would rather have Jim here and for the Grim reaper to have done his assigned job on you.

You are a cancer on this group.

A lot of good people have left because of your continual attempt to be relevant, which you can only do with your mindless attempts to put others down. If you were the man you want us to think that you are, you wouldn't act like a baby cutting their first tooth.

 

[Michael Terrell]

On Sunday, January 26, 2020 at 3:42:40 PM UTC-5, George Herold wrote:

On Sunday, January 26, 2020 at 11:16:31 AM UTC-5, jla...@highlandsniptechnology.com wrote:
On Sun, 26 Jan 2020 09:21:58 +0000, Tom Gardner
spamjunk@blueyonder.co.uk> wrote:

On 25/01/20 21:22, George Herold wrote:
The other bad thing about selling stuff is you've got to
spend ~1/2 your time on marketing... maybe more early on.

And all the other administrivia Even if someone
else is doing all that guff, and engineer operates
at full throttle for about 30%-40% of their time.
The rest is "unproductive" meetings etc. Hence in
the absence of better information, multiply your
first estimate by 2.5-3.5

I've seen a couple of startups make one mistake;
I suspect you are too smart, but here it is anyway...

The founder(s) are trapped by the glitz of being
"self-employed CEOs", and spend money on unnecessary
/visible/ trappings, e.g. cars, boardroom furniture,
and the like.

Some people think that the appearance of success (hence buying all
that glitzy expensive stuff) is success. It's the opposite.

One guy that I know of got some investors to join up, and first thing
bought a private jet. Next thing, they fired him.

(One common mistake is to give up too much equity early on, often to
less than savory/competent partners.)


They feel great and "energised", but none of that
gets money in the door.

More successfully...

I've also seen successful R&D consultancies that
only buy the minimum of equipment, i.e. equipment
that is guaranteed to be useful on every project.

General-purpose equipment is good to have around. These days, a suite
of good test equipment isn't very expensive.


Any equipment that doesn't fall into that category
is bought on a project-by-project basis and after
the project it is part of the deliverable to the
client.

NO! Keep all the toys! Well, you can provide custom test sets at large
additional cost. Price them separately; they may find that they want
several.
Yeah, a bad thing about losing my job, was losing most of my toys.
(I've got some power supplies, rigol scope and rigol sig. gen.
and as you say parts are pretty cheap these days.)

George H.



Good luck; all startups need that.

One can certainly be a consultant, and sell the IP and designs and
equipment on a one-time basis. That gets a company started, and some
people can do that forever. But that requires constant selling, and is
often feast-famine stressful. Selling products is a better long-term
force multiplier.
Right, you need a few horses in the stable for sales (at least at the
low volume level I'm contemplating.) to be semi-stable.

The other thing I think about is fun little pcb's and a part's bag/ list.
There's lot's of rather mundane circuits you can think about,
(gain, filtering)
fun ones are rev. biased led spads, and fast edge TDR's

Some standard box for little circuits would be nice.
Phil, what do you pay for your 'stomp' boxes.
(two piece diecast )

If you need a piece of test equipment, ask. A lot of us have an item or two that we wouldn't miss, and at a better price than the used equipment dealers or untested on Ebay. For my use, I bought mostly damaged or 'for parts' items and repaired the best of a model. Sometimes it was a working but damaged item, and one that looked brand new but wasn't worth repairing. Like a Fluke 8920 True RMS (up to 20MHz) like it had either been used in a High Voltage circuit, or it had been hit by lightning. It did yield a nice case. ;-)

There are a lot of diecast and extruded cases on Ebay for prototypes. There are some nice knockoffs of Hammond diecast with a brushed finish.

<https://www.ebay.com/sch/i.html?_from=R40&_trksid=m570.l1313&_nkw=1590BB+120x95x35mm+Aluminum+Box&_sacat=0&LH_TitleDesc=0&_osacat=0&_odkw=111599975005>

I found some two piece 25mm extrusions that you can use with 'large flange' BNC,N and other connectors with a 1" square flange.I use them to build RF modules and test fixtures. These come in several colors and lengths.

A little blurry, taken with my cell phone

<https://www.dropbox.com/s/vb8ezxigea6an1u/2020-01-30%2013.39.39.jpg?dl=0>
<https://www.dropbox.com/s/q9734ehl1g4vpwj/2020-01-30%2013.39.49.jpg?dl=0>
<https://www.dropbox.com/s/q3vu08mggnjcci4/2020-01-30%2013.39.57.jpg?dl=0>

 

Michael Terrell <terrell.michael.a@gmail.com> wrote in
news:85b07588-82ef-42dc-a6a0-cad3ee2772ca@googlegroups.com:

A lot of good people have left because of your continual attempt
to be relevant,

If folks left due to a bad taste problem the group would be empty
except for you.

Some folks leave... some folks stay... you knowing any real stat on
the matter is pure horseshit... guaranteed.