H. T. Love

Technological Complexity and the End User - A Work in Progress

It just occurred to me that some may be wondering why I'm posting these comments. They are intended to provide a foundation that will let us examine some possibilities that the article does not touch upon. These could affect the outlook we see for the carriers over time.

If most feel I'm wasting your time and my energy, please state so. I will take no offense and do understand that not all need the depth of comprehension as I need to make my decisions.

Regarding "backhaul" as used in your article, see

en.wikipedia.org/wiki/...)

Moving on, you said "The Example of Satellite TV: If you have satellite TV, you are connected into a one-way mesh network" and "All you need to do is to add two way communication through the satellite receiver".

Keep in mind that most commercial broadcast satellites only have a limited amount of "uplink" capacity (only a relatively few points at any time will be transmitting to any particular satellite). If a large number of new ground stations (e.g. a significant number of homes in the U.S.) began transmitting, "uplink" capacity could be overwhelmed. A lot depends on the nature of the "uplink", digital or analog and the capacity (speed) to be provided to each ground station versus the available capacity (speed).

Analog requires division of the bandwidth (the capacity of the signal to carry information in a given time frame - the frequency) into discreet "channels" so that the signals from various sources do not become intermingled. And there is also a "buffer" channel that provides spacing between active channels to allow for the "bleed over" that occurs from modulation of the signal. "Drift" in bandwidth and placement in the frequency range occurs due to temperature changes, manufacturing tolerances, etc., requiring the "buffer" to isolate adjacent signals from each other.

An analog signal is just a carrier signal in a specific band (frequency) that has information encoded onto it. The encoding is done by adding a continuously variable voltage (from max positive to max negative volts added to the carrier voltage in the case of AM - Amplitude Modulation) or varying the frequency within a predefined range (for FM - Frequency Modulation).

"Digital" is a "formatting" imposed on the data which is applied to a carrier signal and also requires bandwidth allocation and buffer channels. Data is carried as a series of "on" or "off" "bits" as opposed to a continuously variable (within limits) signal for AM or FM analog streams. By encoding the data into digital "packets" that carry "control information" (e.g. TCP/IP protocol) and interleaving packets from various sources onto the signal, each band can be shared by many "streams" (e.g. users) at once. Since the amount of data that would normally be up-linked from typical residential users is relatively small (no live video feeds - yet - that require substantial bandwidth even when compressed), the number of "uplinks" that can be supported simultaneously within a band is increased substantially by this packet interleave strategy. Lots of streams of data from multiple sources can share the available band-width.

But even this strategy doesn't get rid of all the issues. Imagine thousands of transmitting points simultaneously sending their packets on the same frequency. There will be "collisions" where the signals and their information content will blend with others and corrupt the information. We can see that we need to get the "uplinks" isolated into discreet bands (by dividing the available bandwidth into sub-channels that use smaller parts of the available frequency) to avoid this and/or have a validation and re-try facility in place to ensure uncorrupted transmissions (this is what normal ethernet uses - collision detection that requests re-try when a collision is detected, see

en.wikipedia.org/wiki/...

for more information).

I think satellites don't have this capability right now. Coalescing and interleaving of packets needs to be done elsewhere or have a protocol added to both the satellite and home unit to "negotiate" the band and capacity and the satellite needs a facility to integrate these disparate streams into an interleaved "downlink" to some ground station(s) that can parcel out the data to the next step on the way to ultimate destinations for each stream of data. The "head equipment" (the equipment currently used to "uplink" broadcast streams) at the ground stations might need supporting capability upgrades.

If I recall, most commercial satellites have some control functions that a technician at a ground station can use to re-allocate bandwidth. This could ameliorate some of the issues as a large number of new "uplink" points came on-line. But I doubt it is sufficient to handle a "mass migration".

On last consideration. Most "interactive" satellite receiver setups require some auxiliary communication facility to enable two-way communication, a phone connection or internet connection through your computer, etc. This is due not only to the limited "uplink" capacity mentioned above, but also due to the fact that it takes a *lot* more power to send a signal up to a satellite than the typical home receiver-antenna (which, by the way is uni-directional now) can muster. There is also an issue of aiming the dish more precisely due to signal concentration into a very focused narrow "beam" (to minimize power required) that will have sufficient strength when it reaches the satellite. Anyway, the implication is a higher power output (possibly subject to FCC regulations regarding RF interference or other issues), higher electric bills, completely different equipment needed for the antenna and receiver (to become a receiver/transmitter) at the home.

Regardless, back to the main issue - do we have a LAN? Further, I add why is it important to know this?

You also said "and now you have a full local area network".

See my prior comment about LAN, WAN and "mesh". This a very complex subject and is made more so by the fact that we no longer define networks simply by their physical connections, but also by their logical connections. For a good starting point see this article about network topology

en.wikipedia.org/wiki/...

which provides links to discussions of LAN, WAN, and many other (to me) interesting things.

Don't be surprised if some of those articles puts the lie to things I've said or will say.

Why is it important to know if we have a LAN after "simply" making our satellite equipment bi-directional?

It is because the capabilities you have are determinwd by whether you are part of a LAN or a WAN. The capabilities, and likely the cost, of your setup will be determined by whether you are part of a LAN or WAN. Also, your personal LAN is not a direct result of becoming a member of either a WAN or LAN via a satellite or any other provider. Your personal LAN requires configuration of equipment and/or software facilities in your local area (whether it be your home, neighborhood or ...).

This affects the cost to the carriers of providing the facilities that may be envisioned. That may affect their (and our) investment and construction decisions and long-term strategies.

In the following discussion, it may help to envision the ground station, satellite and home equipment points connected by satellite signals as being connected by a sequence of "wires" between points with many "wires" to many points emanating from the satellite. The satellite is potentially the "switch" from one wire it "sees" to another it "sees", possibly under the control of a technician at some ground station.

Picking up with the adding of bi-directional capability to your home equipment, what you have is a physical network connection. Whether it is a LAN or WAN connection depends on the scope and accessibility (addressability) provided by the physical connections and/or routing protocols implemented, usually by some configuration of tables of addresses used by software.

Do we have a LAN?

Yes and no. If we presume all the above issues are addressed, we now bump into the scope and isolation issues mentioned in my prior comment. Obviously your node is not physically isolated and it has the *potential* to communicate with another node in a remote location served by the same satellite or ground station - you are *potentially* part of a WAN. You may be part of a WAN beyond the satellite or ground station that provides your physical link. At a layer above the physical, you may be part of a WAN (e.g. established via a Virtual Private Network - VPN) that allows you to communicate "directly" with a remote node served by some facility to which you have no direct connection via your home equipment, satellite and ground station signal facilities.

You may be a member of a possible LAN defined beginning at the satellite (or ground station) that established the connection and addressability of your node. For example, presuming the capability exists in the satellite, you might have direct communication to another home in your neighborhood, via the satellite (assuming it can act as a switch or router for nodes that it "sees").

If you have a switch or router setup at your node that can reach other nodes without going through the satellite, you may also have a LAN. If not, no LAN unless the satellite connection provides it. Your switch or router might enable communication with your home, neighborhood or sub-division over various media and you would have a LAN.

That's all for today. I might do some more in a day or two if you all haven't shooed me away.

HardToLove

Technological Complexity and the End User - A Work in Progress

I've not kept up with all the technicalities over the years, but I figured as I go through the article again over time, I would give a little thought and comment. My first thoughts follow. Any errors of fact or fiction are purely due to antiquity (of myself and my knowledge) and laziness (for now?), not having time and inclination to start digging into all the years of changes that have occurred over the decades. For our purposes this should make no difference.

I decided I would start to chip in here as proper understanding of basic concepts have a big affect on what we think may come about in the future, affecting our investment decisions. Keep in mind that there are some /very/ bright people out there and forecasts we make based on our (incorrect) understanding may end in disappointments.

You state "In a mesh network more components of the network are interconnected wirelessly".

That's only a decision for added capability, convenience, cost, difficulty of implementation or maintenance reasons, and/or speed of deployment. Limiting factors of "cable" might also affect the decision. Regardless, it is entirely possible to make a "mesh" network based on cable. It's also important to remember that "networking" incorporates multiple layers and that "meshing" (and networking itself, in fact) must occur at the physical layer and may be enabled or disabled at a logical layer.

Mesh networks can be constructed using "cables". An application using "cables" might be an installation wherein ultra-high security, availability and reliability are mandatory. Another might be where existing cable-based infra-structure already exists and added functionality or increased availability of all functions and features are desired and the "best" solution (least cost for the results) includes using the underlying infra-structure and augmenting it.

Mesh networks can also exist underwater and use neither cables nor radio waves.

You also state "A mesh network is also called a LAN (Local Area Network). Less cable, less cost."

For the "Less cable ..." part, see above.

A mesh network need not be a member of a LAN and need not be a LAN. To call it a LAN is correct only insofar as a particular implementation has the attributes of a LAN. It could be a member of a wide-area network (WAN) or be a WAN itself. A LAN could be part of a meshed network and could also contain a meshed network. It could also have access to a WAN or another LAN through one or more members of a mesh network.

LAN and meshed network are two totally different beasts. LAN and WAN are only about scope and accessibility while meshed networks are only about dynamic detection of physical nodes and dynamic construction of a network using detected nodes.

LAN (Local Area Network) has to do with localized "scope" (local-only addressability) and isolation, as opposed to a WAN (Wide-Area Network) which has global scope (global addressability) and integration, like the internet. One of the most frequent implementations of a LAN includes (used to include?) physical isolation (no physical access to a WAN). These days many LANs, but not all, are provided access to a WAN via one or more access points on the LAN that interface with the WAN. All LAN<->WAN traffic must transit these access points.

"Mesh" is all about dynamic flexibility of making connections and routing. This may be used in a LAN or a WAN and may be used to implement a LAN or a WAN or connect LANs and/or WANs together.

The concept of LAN and "mesh" is not tied to particular media. They could be wired. That would certainly make a mesh (pun intended) of the wiring, but meshing could be done that way.

This /may/ be a significant consideration in that any accessible physical-layer (radio waves, copper wire, fiber optic, low-frequency long-wave deep water sounds, ...) access point that has bi-directional capability and a routing protocol (could be hard-wired but is usually done via software and dynamically constructed tables these days) could potentially become a node in a dynamically constructed mesh.

Here I started my usual "brain dump" mode with a full history of how a need for definitions of things like LAN and WAN came to be. Fortunately, I "came to me senses" (my ancestral Irish brogue creeps in occasionally) and have deleted all that.

So I'll end here with examples of mesh networks that either are in use today or could be used for certain needs.

- Battlefield C3/C4 - wireless,
- Submarine communications - long-wave undersea sounds
- Super-secret security-hardened installations that must emit no detectable EM (electro-magnetic) radiation, have redundant network management and access points so that failure of one node doesn't affect others - cable based (probably fiber-optic),
- High-availability, high bandwidth server farms and campus networks - multi-media.

In each of these cases, the mere appearance of another node (whether by plugging in a cable, activating a radio-frequency transmitter/receiver, emitting/detecting sound waves from an undersea communication point - SOSUB? - or any combination of these) could invoke a "meshing" operation if the software is in place to accomplish that.

I'll comment more later.

QUICK CHAT #30 Start 12/18/09 after Market Close/Bernanke Villian of the Year

On Dec 19 05:47 PM tripleblack wrote:

> Dean still harbors dreams of the Presidency, and one way to open
> up the 2012 Dem slot would be if ObamaCare (synonymous with Obama
> in lots of folks' minds) went under. I doubt his motives, is what
> I am saying.

Regardless, he struck me as the best form of Dem - fiscally and operationally more conservative than most Dems in power and socially liberal. Need that second part to differentiate from a Republican (um, come to think of it, the first trait too!). The important part to me is that he had the cognitive ability and drive (guts?) to publicly identify the real issues with the Dems process and results even if it might bite him later on. That implies a "scrapper" willing to go against the grain when it seems the thing to do, regardless of what his motivation is.

I like that in a person. Disclosure: might be some empathy there as it has been one of my predominate behaviors over the years, sometimes to my chagrin.

> <snip>
> Its INEVITABLE that there will be a landmark heathcare reform bill
> with Obama's name on it. There is no escape. It is not a question
> of IF, or even WHEN (obviously, not long from now), but WHAT is in
> the thing.

He said that he has actually read the whole thing! Another trait I admire - willingness to roll up one's sleeves and do the grunt work.

> <snip>

HardToLove

QUICK CHAT #30 Start 12/18/09 after Market Close/Bernanke Villian of the Year

On Dec 20 03:19 AM Freya wrote:

> Paul refuses to buy an Intel CPU, he started using AMD years ago.

Smart man! The /only/ Intels I have are an old 200Mhz Pentium (but that's "offset" by an AMD K6-III) from the days when I was "in the biz" and the one that came in my Toshiba laptop (the price, cheap extended warranty and features overcame my preference for AMD).

>
>
> His commentary has to do with Intel's greed. They seem to introduce
> a "new" chip almost every 6 months and each new chip has to have
> a New MoBo.

Mine has to do with their lack of innovation at a time when AMD was making faster performance with fewer clock cycles, longer instruction pre-fetch and decode pipelines, bigger level 1 cache, reduced power consumption, and many other advances. Intel just kept bumping the clock speed, eating more power and generally lagging the curve.

>
> The newest cases don't require tools, you can pretty much install
> or replace anything without going to a Geek. But he has always been
> leery about removing and replacing a MoBo.

That takes the most time and is generally a PITA (Pain In The A$$). But a simple # 6 Phillips screwdriver, attention to alignment with the case expansion slots and getting support bosses everywhere you can is the key thing.

> <snip>
> In 2010, the Chinese will descend on Intel like a Plague of Locusts.
> A new chip with a new architecture, not 86xxx based, called the Godsend3
> ( not positive on name but read about it early this year ). RISC
> based, software compatible with 86xxx. Its supposed to be the 3rd
> generation of this architecture and has been unavailable outside
> of China. World release date was to be 2010.

IBM, and others, had RISC chips long ago and couldn't overcome the Intel market share. But they didn't have the burgeoning Chinese market that exists now. The PowerPC was also a much better CPU, especially in the I/O handling, but also couldn't get the market share needed.

>
> AMD has beaten the pants off of Intel on the Graphics side,

Uh, actually that was Nvidia and ATI, which AMD recently bought. Selling their foundry and buying ATI was one of the better moves they've made in recent years. It may put them in position to compete more effectively against Intel in the future. The market thinks so too, based on share price over the last year or so.


> Godsend3
> might outdo Intel's cpus. The Chinese do not want to be dependant
> on the West for anything.

Actually, I think with the Chinese market growing so much, the result may be much more startling. As with PMs, we may become dependent on them in this area as well. Our government doesn't support home-grown innovation and manufacturing like it used to, so our local may again get crushed by this.

HardToLove

QUICK CHAT #30 Start 12/18/09 after Market Close/Bernanke Villian of the Year

I think you made a good buy (I hope so!). My wife and I have the 17" screen (biggest offered at the time) for the same reasons.

A tip: a larger screen can be set to lower resolution and still have a lot of "real estate" on-screen. By default, I think, they are set to the highest, these days probably something along the line of 17xx by 12xx pixels.

Play around with your settings (maybe something like a 1240 by 1068?) and you get larger letters and symbols and still get a lot on the screen. Then set the browsers et al sizes to suit yourself and you're good to go.

That's what I do on all my 'puters.

HardToLove


On Dec 19 04:46 PM optionsgirl wrote:

> I bought a new laptop. Toshiba w/ a 18.4" screen ( that's even bigger
> than my macbook pro! Very important because the older I get the bigger
> the print has to be, so if the screen is small, it's a challenge...)
> it's blue tooth compatible and the three year service contract is
> really like insurance-it covers accidents- so if you drop it- you're
> still covered- unlike aapl, which voids the applecare if it is damaged
> by the owner! Price was $845 plus $300 for the service contract.
> I forget, I think it has 6 gigs of memory, and you can upgrade it
> to 8 gigs. Does that sound right? My eyes cross while these guys
> talk, my husband asks the questions, I just plug it in and stare
> at screens all day. YH Best Buy is selling that Kapersky (however
> you spell it) protection for $59 for a year.

QUICK CHAT #30 Start 12/18/09 after Market Close/Bernanke Villian of the Year

Indexes re-balancing.

HardToLove


On Dec 19 04:47 PM ScroogeMcduck wrote:

> There was some pretty heavy volume in the S&amp;P at the end of the
> day, also.

QUICK CHAT #30 Start 12/18/09 after Market Close/Bernanke Villian of the Year

I'll tell you what - it's got to be bad. Howard Dean, a very smart Dem regardless of his politics and hi-jinks, says it stinks to high heaven, doesn't accomplish it's goals, etc. He said he would not vote for it. He wants to strip the bill to the bare minimums and than augment over time as experience and reality dictate.

Check CNBC for the interview.

HardToLove


On Dec 19 04:55 PM tripleblack wrote:

> www.foxnews.com/politi...
>
> Reid announces he now has the 60 votes needed to pass ObamaCare.
> Not sure yet on how much baksheesh pork they are heaping on Nebraska,
> but the grain crops will probably be amazing next season!
>
> Looks MORE likely (who knows how the final bill will shake out) that
> there WILL be some sort of private health insurance business, long
> term, but that its profits will be regulated very tightly.
>
> Going short the health insurance industry (do they have an ETF?)
> might be a good play right about now.

QUICK CHAT #30 Start 12/18/09 after Market Close/Bernanke Villian of the Year

Found this

en.wikipedia.org/wiki/...

HardToLove


On Dec 19 08:44 AM optionsgirl wrote:

> Does anybody know what a wi fi mesh network is and how this could
> be good for T and VZ?

Cramer's Stop Trading! Buy the Heck Out of Citigroup (12/17/09)

As counterpoint to the sentiment "these losers", guys like me tend to think of them as hapless. It makes me sad to think that some of his audience might lose most of what they have sometimes.

Some may not have time for doing their homework, as he always advises, or other good advisors they can rely on.

HardToLove

On Dec 18 09:53 PM Dialectical Materialist wrote:

> Cramer is valuable... Every trade needs someone on the losing end
> and he restocks the pond with these losers every time he recommends
> something.

QUICK CHAT # 29 Thursday 12/17/09

Between my wife and myself, we've had excellent results with Toshiba. Ours had 3 year warranty available for the most reasonable price I've seen (never had to use the warranty). My wife's from 2003, mine from 2004 still going strong.

My daughter has Dell and had to exchange several times over the last couple years.

HardToLove


On Dec 18 03:38 PM optionsgirl wrote:

> I am going to buy a new laptop this weekend. Any recommendations
> (other than Dell)?

QUICK CHAT # 29 Thursday 12/17/09

If it's one of the newer ones, it's probably because thy switched from PowerPC chips to Intel. New bus, lower I/O performance, more "commodity" parts, etc. Chasing market share through reduced cost. Reliability and performance can suffer.

HardToLove


On Dec 18 02:09 PM ScroogeMcduck wrote:

> OG thats disappointing news I have had a G4 since 2000 and its still
> kicking my new Dell's butt (with a few memory upgrades anyway)<br/>What
> seems to be the problem?

NVAX: Interesting charts - UNDER CONSTRUCTION BUT NOT A LOT TO ADD

I've made some $ playing the swings. Since long-term, to me, looks so uncertain on so many things, I'm just scraping the table, around my core holdings, for returns while I wait for Act 3 to appear and play out.

Today I've got a trailing buy stop in, that if it does it's normal rise I should make a few hundred selling later in the day. If it gets too late and the buy hasn't triggered, I'll cancel it and probably wait for another day or I'll manually work it.

HardToLove


On Dec 18 12:28 PM Mark Bern wrote:

> YH - I think you're on the right track with this one. Lots of potential
> over the longer haul, but the short-term view looks a bit too volatile
> for me. I like volatile, but only when it is in the other direction.

NVAX: Interesting charts - UNDER CONSTRUCTION BUT NOT A LOT TO ADD

I made it in a spreadsheet from data available @

www.nasdaqtrader.com/T...

Be aware that like the SEC failure data, it runs late and is "batched" up - no daily/weekly ...

HardToLove


On Dec 18 12:25 PM Mayascribe wrote:

> HTL: Great chart. Where did you get hold of the short interest chart
> for NVAX?

NVAX: Interesting charts - UNDER CONSTRUCTION BUT NOT A LOT TO ADD

On Dec 18 12:18 PM User 283977 wrote:

> Good data HT.... can you develop a chart that has number of shorts
> on say the X axis and the NVAX price on the Y axis? That would let
> us see both sets of data at the same time resolution, and to clearly
> show the nature of the association between number of shorts on NVAX
> price. I am accumulating NVAX.

I considered that, but was in a hurry. I had already taken a snapshot for the 1 year price chart and then accumulated the data for the delivery failures and short interest. I think I can actually put them together using GIMP on my Linux platform and want to give it a try and see how well it works out. It depends on if the top layer transparency and the bottom layer background leave things clearly visible and don't give too much clutter.

On difficulty with putting both in the spread sheets is time-frames. Short/fails are in two week intervals and prices, of course, are a range - O/C/H/L and daily or weekly or ...

The source of the fail/short, if I recall, has some price but we don't know if it's VWAP or what.

I will do some more looking and thinking, but I did want to get these up today.

By the way, "click to enlarge" and you not that although they are not perfectly aligned, the time-frames are close enough for what I was trying to accomplish - see the rise in prices start up followed right on its heels by short interest.

HardToLove
>
> I am impresed with your use of the SA chart tool. I found it to
> be very difficult to use.

QUICK CHAT # 29 Thursday 12/17/09

ANDINA Minerals ADM in CA, (ADMNF) in U.S.

Just found, no DD yet, but news found today. Web site @

www.andinaminerals.com/

... today announced initial results from the strategic exploration program at the Ojo de Agua East gold target, located at the Company's Volcan Gold Project in Chile. ...

Since tables of gradings included are hard to format in this comment format, I chopped the rest. But the full text is available through the headline on the web site home page.

Thought I would pass it on.

HardToLove