That marketing discussion
potatohead
Posts: 10,261
You know what?
I'm not fighting that battle. I've a good pile of stuff written up, and I'm going to file it.
Why?
The hassle isn't worth it. Truth is, this is a highly differentiated device as the P1 is. If we can't speak up to those strengths and identify it's parts in equally differentiated ways, the whole affair will be doomed to comparisons, and where that's true, I don't think we have a winner in terms of messaging.
Secondly, there is a conflation of things!
How education presents it's material is one problem. Common, historic development lexicon is another. A third is the message overall, positioning, etc... Yet another is the simple problem of getting the device listed.
Unifying these things makes no sense. Been there, done that kids. Generally, it's a fail.
I'm not going anywhere. It's not that. Love the community, love the development going on, and I love the tech. However, I don't feel good AT ALL being asked to manage my speech here, ON THE FREE GIG MIND YOU, toward some end I really don't feel good about, so I'm not going to, and we all can still be friends.
I've not had one lick of trouble with COG and HUB and educating people. The parts of a Propeller need names that can be associated with the attributes and dynamics between those parts too. We have that, and it works. It could be improved, and it needs some additions given the new design. Going generic isn't the cure all.
I've been communicating technical information to people for 20 years and personification along with differentiation well applied carves out the necessary bits where the magic is, while more generic lingo works to link them to other experiences, skills, etc... That formula works, and I don't feel good about not doing it at all. Runs counter to my nature and experience. I can count the complaints and general failures on one hand, despite hundreds of diverse experiences.
The net of the discussions I had over the last day or two with some peers I've known for a long time was interesting. A primary and common thought was some basic goals need to be established in the various compartments I mentioned above. We have a little of that, but more is needed.
In the past, I've struggled with education driven by marketing for example. Hated it. Still do. Always will. There is some need to do the work to bring the terms into more familiar contexts, but there isn't a need to make everything generic. Been there, done that too. How you sell it has very little to do with how you educate people about it.
Additionally, where there is history, there will be terminology. Those references are invaluable, and redefining things invalidates them, meaning we shrug off a lot of meaning, experience, community, etc... that has huge value. I won't be a part of doing that. It's Orewllian. See below.
An example from my recent experiences is worth it here. One of the 10 largest companies in the world attempted this discussion and completely failed. They had the same desire, and that was to unify all the modes of communication to maximize the messaging. And they failed. They even went back and edited, deleted and modified older discussions, documents, etc... and they failed big. Got a shock when people simply republished and carried on without them.
Truth is, those people are using the terms that grew organically, and they are leveraging their user base, newbies being mentored in, and the result is a nice blend that works. Document after document comes out, and it gets edited, people say what works, and a lot of time and energy is wasted. I want no part of that here.
In every other way, it's all good. Have fun. I am, and having fun is a part of why I'm simply not going to entertain the discussion any longer. This might be one of those "bell rings at the school of hard knocks" kind of things. I'll be around when or if it rings and we can talk then.
I'm not fighting that battle. I've a good pile of stuff written up, and I'm going to file it.
Why?
The hassle isn't worth it. Truth is, this is a highly differentiated device as the P1 is. If we can't speak up to those strengths and identify it's parts in equally differentiated ways, the whole affair will be doomed to comparisons, and where that's true, I don't think we have a winner in terms of messaging.
Secondly, there is a conflation of things!
How education presents it's material is one problem. Common, historic development lexicon is another. A third is the message overall, positioning, etc... Yet another is the simple problem of getting the device listed.
Unifying these things makes no sense. Been there, done that kids. Generally, it's a fail.
I'm not going anywhere. It's not that. Love the community, love the development going on, and I love the tech. However, I don't feel good AT ALL being asked to manage my speech here, ON THE FREE GIG MIND YOU, toward some end I really don't feel good about, so I'm not going to, and we all can still be friends.
I've not had one lick of trouble with COG and HUB and educating people. The parts of a Propeller need names that can be associated with the attributes and dynamics between those parts too. We have that, and it works. It could be improved, and it needs some additions given the new design. Going generic isn't the cure all.
I've been communicating technical information to people for 20 years and personification along with differentiation well applied carves out the necessary bits where the magic is, while more generic lingo works to link them to other experiences, skills, etc... That formula works, and I don't feel good about not doing it at all. Runs counter to my nature and experience. I can count the complaints and general failures on one hand, despite hundreds of diverse experiences.
The net of the discussions I had over the last day or two with some peers I've known for a long time was interesting. A primary and common thought was some basic goals need to be established in the various compartments I mentioned above. We have a little of that, but more is needed.
In the past, I've struggled with education driven by marketing for example. Hated it. Still do. Always will. There is some need to do the work to bring the terms into more familiar contexts, but there isn't a need to make everything generic. Been there, done that too. How you sell it has very little to do with how you educate people about it.
Additionally, where there is history, there will be terminology. Those references are invaluable, and redefining things invalidates them, meaning we shrug off a lot of meaning, experience, community, etc... that has huge value. I won't be a part of doing that. It's Orewllian. See below.
An example from my recent experiences is worth it here. One of the 10 largest companies in the world attempted this discussion and completely failed. They had the same desire, and that was to unify all the modes of communication to maximize the messaging. And they failed. They even went back and edited, deleted and modified older discussions, documents, etc... and they failed big. Got a shock when people simply republished and carried on without them.
Truth is, those people are using the terms that grew organically, and they are leveraging their user base, newbies being mentored in, and the result is a nice blend that works. Document after document comes out, and it gets edited, people say what works, and a lot of time and energy is wasted. I want no part of that here.
In every other way, it's all good. Have fun. I am, and having fun is a part of why I'm simply not going to entertain the discussion any longer. This might be one of those "bell rings at the school of hard knocks" kind of things. I'll be around when or if it rings and we can talk then.
Comments
???
Not sure what your complaint revolves around completely, unless it is the talk about trying to help Parallax sell products by making them easier for potential customers to relate too.
The P16X32B really seems set-up to SLAY a large portion of the AVR/PIC/ARM competitors based on its multicore, 256/512K RAM, and 512 Register makeup. The 2014 P2 became too complex for a 180nm process that had its peak in 2000-2001.
Parallax needs something to market in the near term, and that something has to be 'comparable' to all the other somethings everyone else uses and understands. At least thats my take, since they went to the effort of starting Parallax-Semi...
Whether the P2, or the P16X32B, I think if Parallax were to once again try to sell the whole Propeller Cog & Hub thing, it would fare somewhere in the same part of the Bell Curve as it has in the past. Or actually, less, as Cortex M0's are down to what $0.50? Doubt margins would be as great, which makes it imperative that they sell more of the P1, or they raise the price, and have a harder time competing against cheaper, 'good enough' ARM's.
Where you see Orwellian, I see it more as experience learned from the P1 'differentiate yourself' attempt.
P1 differentiated way too much. This time, it looks like they are differentiating only moderately at the surface, and much more in-depth if the customer so wished.
EDIT-
I think I get whats got your goat.
"Truth is, those people are using the terms that grew organically, and they are leveraging their user base,"
From my understanding of some of Ken's comments, Props user base is academic, and some commercial, and forumistas.
Is the revenue ration 80/15/5, 50/45/5, 80/18/2 ? Only Parallax knows for sure, and we're left to conjecture.
I would say however that most of the education market seems to be one-offs, from a student perspective. How many forum posts have you seen that mention the poster had started with the Prop in school? Outside of homework or school project related posts?
My point is you seem to be lamenting a potential change in Parallax's marketing of the new Prop as Orwellian or abandoning some large user base.
Outside of the forumistas there are x number of commercial developers, and some number of educators.
I think the forumistas and the commercial dev's are following along fine, and many will find this a commonsense move.
The educators may or may not have to adjust lesson plans, however they probably like this as it makes their lessons more germane to the industry at large.
I understand nostalgia, however is it really fair to Parallax, its staff, and the revenue they may generate by potentially being more successful ?
----
WARNING- This poster's comments are known by the State of CA to potentially contain sarcasm, wit, and logic.
May cause chip failure if over-clocked. Please read with compiler settings at -o2 -reading-comprehension
Let me add this: Weaving what we've got together with a more potent message is going to be a considerable effort to do right. That effort shares some space in common with education, but that one has some uniqueness to it that runs orthogonal to selling type activities.
This effort done with some goals, research, planning and recognition of the different domains and disciplines may well result in cog, hub, whatever being eliminated from the sales messaging, and that's fine. Education has some different needs, and we either republish a lot of stuff, or we forgo it, or we weave it together better and apply techniques to optimize in various parts of the world, etc... Taking it entirely out of education is more expensive than it seems at first blush.
As for historical lexicons, those tend to endure no matter the amount of money spent. Again, a top 10 multinational, global company has failed in this. I've seen a couple in the fortune 1000 stumble hard too. In both cases, their own people using finger "quotes" where needed to bridge gaps created by failure to recognize all the dynamics, and doing that is expensive.
I would much rather just do my thing and let the results happen as they will. Maybe that bell rings at the school, or not! I hope not, as it's much easier to join a successful effort, which I will easily do.
For me personally, and this is not aimed at anybody in particular, I either get sucked way in and contributing will be a project, or I'm content to just let it happen. I'm sold. No need to worry past that. So it's about me, let's be clear on that.
Like I said, we can be friends. Some of us may jump in and do great. Parallax is going to have to noodle on this, ideally investing some to really make it sing, but maybe just doing their best too. It will be what it will be, and I just don't think my experiences and nature align well with that, so it's better to step away and do what has value. Again, about me.
My intent isn't bad, and I'm not assigning any to anybody, just making a few things clear given what I had posted here, that's all.
You get a gold star!
I'm kind of itching to get an FPGA to start on some TV drivers I want way more than I want to work on the marketing, and I'm either in on that, or out. Out it is!
To be honest, I'm not sure the discussion needs to happen right at this time, anyway. There's a lot going on.
Who knows? If you all are really clever, I could get sucked back in. You never know. I'm not angry. Just needing to draw a boundary.
I keep some ideas ticking along and hypothesis-test them from time to time, but its hard to get the proper perspective, being 'inside' and already knowing and caring about the propeller family.
Lets keep making notes and discuss at a better time.
Actually that's not entirely it.
Some of it rings true. Nostalgia is appealing.
One, I really would need to own it, and I'm not in a position to do that. And that's fine too. I hold no expectation of being in this position either.
Two, I really believe in strong product differentiation and a whole lot of the generic "standard" messaging I've seen tends to educate fairly well, but it also invites way more comparisons than I would like, and it minimizes some core differences that really matter. It's those things we are selling! And if we can't stand up and be proud about it, then the selling won't work all that well. And that whole discussion is a mess, because we are mostly geeks, bunch of us nerdy too, and that's great! Love it here, and I wouldn't have it any other way. But that's not gonna get it done.
Coining terms and doing the work to make 'em stick is a big part of this. The real art is expressing them in tandem with the generic so that they stick out and people go, "so that's it!" Pure magic one doesn't get the advantage of, unless this is understood. Making it happen is a significant amount of work, work I won't / can't do unless there is backing for it and a goal to measure it by. We lack both.
Three, the idea of being a niche is regularly associated with doom, or failure, or inadequacy. None of those things are actually true! We have had that talk before. Parallax needs more. But more relative to the market as a whole is not much. Parallax could double, and be seeing wild success by any metric, and it would be a rounding error in this space overall. I don't think that's well understood.
That has implications on how things are positioned, and sold, if there is even any active effort in the latter area. There should be, but there hasn't been in the past outside of education, from what I understand.
Four, I'm troubled by the melding of what would mostly be internal development type discussion being public and the impact that has on this marketing discussion. Great teams use a lexicon that makes their team great, and that may have little to do with the messaging and education lexicon. Put simply, let's take an easy thing: Instructions and or SPIN + PASM syntax.
We've had calls for standard this and that.
Ok fine, but I didn't come here for standard this and that. SPIN+PASM is awesome, and it's awesome precisely because it's not standard. Fixing that, lets out the secret sauce. Not OK. Instructions, names, concepts used in discussion here by Chip and those of us contributing have similar considerations and implications when diluted down, or mashed into something less expressive.
We are adding standard, and that's great! However, it's not so great, if we dilute the magic, so to speak.
Maybe that helps. Hope so. I'll leave it be now.
I don't want the standard device. I want the one Chip would build, and I want it maximized with the input from those of us here who get it. The language contributes to that, and it's this I feel the most passionately about. It's kind of a freak show as Chip said, but that is a good thing, and it has nothing at all to do with the selling of the product of it. Failure to grok that is something I just need to step away from and hope it goes well.
BTW: Where we've suggested things, sometimes complicated things and we've used generic language to get at them, we've generally expanded beyond what would be simple and awesome. Chip generally factors it down some, and that's been good. Altering the lexicon we use to discuss the vision for the device would lead to losing a few of the magic tricks Chip might bring us, or that we may suggest. Why do that?
Unfortunately, truth is that Parallax does not have the resources to educate people with new terms. They tried this and it failed miserably because you cannot find the Prop on DigiKey etc easily.
You have to use terms that people understand. Then you can say why your product excels or does it differently. But if you never get your foot in the door, it doesn't matter what you call it.
I never use terms like cogs, hub, etc when describing the prop. I use cores, common or shared memory and the cores own memory. Once they are hooked, I might say they call them cogs, hub and cog ram - but I might not either. I prefer to describe it in terms they can understand and then say what else it does, and does better. I love the prop terminology and the beanie, but it does not cut it in the real world, at least until you make it big. Noone worries about the multi-coloured apple these days, but I for one thought the Apple was a toy until someone took me aside and showed me their financial modelling running (3EZY Pieces or SuperCalc) for about $2000 when the minis financial modelling software was >$25,000 and ground the mini to a halt.
Just my 2c.
At first pass it has a tone that implies you have been deeply offended by person or persons unspecified on the forum. Was it me?
But the guts of the message seems to be about changing a little bit of Prop terminology when presenting the PII to the world. About whether that is a good idea or not. Whether to call a spade a "spade" so that people know what you are talking about.
The is no Orwellian big brother. (By the way which one is big brother Chip or Ken?).
It's just a debate right?
I just realized the level of effort required to really participate and have it be meaningful exceeds what I consider fun here, and that I don't want to go there. The rest is why I think that is.
So then, I either just ignore it, which I'm going to do, or I would then have to invest considerable effort to back up my other contributions so far, and that's the not fun part, somewhat based on conversations I had trying to boil down my differences with others in the overall approach advocacy seen here.
It's just going to play out how it plays out. If I'm not really invested in that, then I've no worries. And maybe it plays out just great! Which means I would have avoided a big hassle. If it doesn't, then perhaps discussion at that time would play out differently.
As for unusual terminology, in marketing you always want to differentiate yourself in clever ways. But if the paradigm shift is too great, it confuses the buyer, and they lose confidence that they'll understand the rest of the platform. A new product should have maybe one or two new-fangled concepts, with the rest known and comfortable to everyone.
It's always a mistake to center an entrenched product class (that includes MCUs) around a wholly new idea. If people don't get the new idea right off, then they don't get the product, since the two are so intertwined. You lose sales. Special features are the icing, not the cake.
The core problem here is that the Propeller was made from a somewhat opposite perspective than other microcontrollers. People first need to understand the 'whys' and 'whats' of that, and then they'll have the framework in which to assimilate the details. Therefore, I think it would be cleaner to start off with unique terms that are not pre-loaded and then use simple explanations to describe them.
I get really stressed out about people not understanding things, or not wanting to. I've come to the slow, sorry conclusion that people only know as much as they want to know, and no more. Some people, a minority, are open to new ideas and they are always learning. Most everyone here is like that. It might be futile to try to incite that in others. I heard this somewhere: "A man convinced against his will is of the same opinion, still." Even if you convince one of them of one thing, their attitude won't change, and they'll typically wind up denying what you first convinced them of, as they regress back to their initial mindset.
I've noticed lately I've been getting more and more wrapped up in the forums and it's more and more for less and less technical reasons. The emotional involvement and investment has also increased. Some of it starts out with fun banter but then becomes intense or too time consuming. As a hobbyist/forumite, the more invested I become in some of the discussions, the more it feels like work and the less fun it is. This is my release place, my place to have fun because work for me isn't much fun.
I keep trying to retreat and do more with the Propellers and talk less about them (I'm sure many of you would appreciate fewer comments from me). It is fun to play with the P1 and with the various FPGA emulations and that work is entertaining and hopefully meaningful in some small way. Discussing marketing and such isn't anymore.
I hope for the best success and outcome but I can't directly contribute to it. I will continue to use whatever Parallax offers because it is visionary and fun. I will probably find things about the follow on chips that I wish had been done differently but they will be what they will be.
What he says:
I picked Libby up at the "Peoria International" airport last night. On the first half of the trip home she described how different it was to practice medicine in a capitated system.
Then she asked how my work was coming… I talked to her all about the new FPGA board and Propeller development and after about 10 minutes she said "what is a Propeller?"
After about 10 minutes more of my talking… she said "what is a micro controller?"
We have had basically the same conversation for the last 5 years:)
“A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.”
Exactly. Don't even try. Ask yourself, are you wanting to educate the world as to a nice and easy and correct way things can be done? Or are you wanting to sell stuff and make a living and keep people employed.
No, don't answer. I think I already know. It's a bit of both. You can't do the former without the latter after all.
So, my conclusion is: Don't expect to change the world model of all those potential customers out there. We are not going to live long enough to see that happen. Sell them stuff, with "hooks" they understand and unique features they will appreciate.
A girl friend of mine, many years ago, was really into drama and literature and arty stuff. At the time I was building computers with the old 8 bit chips we had. Sometimes I would try to explain the weird HEX LED display and keyboard and how I could enter instructions and get the thing to do stuff. It seemed to go in one ear and out the other.
Life moved on, we split up, but years later she had a degree in Mathematics and moved to Germany to program the anti-lock braking system on trains.
When we met up and talked about life, she said "you remember that old 6809 computer you were building? That's what inspired me to do all of this".
At the end of the day, the Propeller "approach" will never be the same as the other multi-core approaches, regardless of whether we say "cog" or "core". And that will distinguish it from it's competitors. If we focus on getting the product into the hands of potential users, I think this product itself will take care of convincing those users of it's power and worth.
Some of us are campaigning to change the Prop terminology a little bit so that the rest of the world knows what the hell we are talking about.
That does not change the Props Unique Selling Points, (USP, an apt marketing term), it only frames them in familiar terms.
+1
Relate
Relationship
Customer Relations
I get why Chip, potatohead and others may see redefining terms as 'selling out', or giving a sub-par definition to a feature.
IF the desire is to have a P16/P2 as successful as the P1, then simply continue to market as before.
IF the desire is to have a P16/P2 be MORE successful than the P1, then Parallax has to accept that its previous Marketing was a great experiment.
In many ways the P1 beat the pants off of many of its competitors, from a Technical standpoint.
The simple question one has to ask is, that being so, why did it not generate far more revenue, uptake.
In the same vein, the P16/P2 will be even more advanced from a Technical standpoint.
Since Technical prowess seems NOT to be the determining factor in past Propeller success/uptake, why now would that suddenly change?
I think, from the hundreds/thousands of threads on the Internet, that this comes down squarely to 'relateability'.
You just said the following in another thread:
This is more than just changing terminology. I believe that this very much supports the premise of my statement.
But that doesn't really matter. The point of my earlier comment is that it's okay if we change things about the propeller to make it more familiar to the larger world (even if it's just a change of terminology), as that does not diminish what the propeller actually is! The propeller (and kin) will be successful on its own merits. Our primary concern should be getting it into the larger community. And if that means using more familiar terminology, then we should do that! Thinking about this in terms of Geoffrey Moore's Crossing the Chasm, Propeller users up to this point solidly fall into the "innovator" and "early adopter" part of the curve. Now we need to attract the "early majority". Their needs and expectations are different than ours. If you can get them on board (successfully crossing the chasm), they will be the ones to tell the rest of the world why the Propeller (and kin) should be used.
(note: if there are people involved in this discussion that have not read Crossing the Chasm, I strongly suggest you do that right now.)
Yes, I said that thing, and that other thing, and they are consistent.
Yes, this is more than changing terminology. It's not me, the chip has changed. A PII is not a P1.
For example, talking about LMM, made sense in the PI. That got optimized away into HUBEXEC for the PII. Nobody seemed to notice that "HUBEXEC" is a stupid way to describe it. Executing code from RAM is "normalexec" in the rest of the world. They don't even have a name for it.
Great, we all agree.
Now how to convince Chip and Ken of this...
We have to get the masses to understand and they know the basics. So we use the basics to describe the P16... and then tell them all the other great things it can do.
We have to get them to look first, and that means using familiar terms to them.
I write about this stuff, and I virtually ignored the Propeller until about 2011, when MattG and Ken wrestled me to the ground and stuck some Prop boards into my face. Before that all the talk about cogs and hubs dazed me, and being a busy (not lazy) professional, I opted instead to skip it. I wrote about things I more instantly understood.
OK, I admit I was WRONG, and I was missing a good thing. But here I am, PAID to research these things, and I gave the product a pass. It was marketed with too many strange buzzwords. Cogs? Hubs? These made no syntactic sense to me. I could have easily understood "It's like having 8 microcontrollers in one!" The inherent benefit of that feature isn't lost on anybody.
As a consumer, I'm far more likely to try "Dissolvo, now with Extra Strength Doohicky Bubbles" if the product is presented first in a way I am familiar with. Dissolvo, we understand, is a magnificent product that will rid the world of dirt, gangsta music, and politicians. But if its marketing is too far afield, it doesn't matter how good it is. The product will die because no one will buy something they don't understand.
I like the concept of things like "Smart I/O," but it's not a way to brand the product. Terms like these are meaningless without context, and context takes time to develop for a consumer. Instead show what it can do. Show me the benefits of Smart I/O. Show me an example of a pin-based state machine that with minimal code does what it would take an Arduino dozens or more lines of code.
When we describe the I/O, we should say something like, all I/O pins are equal (no more worrying about what pins are restricted to what, nor conflicting blocks that only map to certain pins leading to a loss of peripherals). All the pins have ADC, pullups, pulldowns, etc. They can be accessed by all 16 cores. Each Pin also has Smart I/O - a small state machine that can be configured by any core to enable the pin to perform complex functions autonomously.
Each of the 16 Cores has identical 32bit micro-controller CPUs, together with 512 32bit Registers organised as 128 x 128bit, accessible as Long and Quad Long, and can also double as private instruction memory for executing extremely fast deterministic code, in addition to executing from the common 512KB Instruction/Data RAM. The 512KB Instruction/Data RAM is organised as 32K x 128bits and can be accessed as Byte, Word, Long, and Quad Long (128 bits).
In specifications and technical documentation you can talk all about specific details, and describe -- by drawings and analogy -- how the architecture works. The latest shopping cart pages for the Propeller now take this approach, so I believe Parallax already understand all this. The pages for the various Prop packages never mentions the terms "hub" and "cog," even in the details section. It's probably in the PDF technical description, but the main pages are all about key benefits that are much easier to assimilate: "Multicore processing simplifies designs," "Built-in Spin language is easy to learn," and so on.
So, this is a different approach than they once had, where they felt they had to describe what made the chip different in order for people to appreciate its value. What I've been reading in some threads here is a throwback to the old method of describing the technical features of the Propeller, instead of showing off its benefits.
I don't check the Forum as often as I should - but I didn't want to pass up this opportunity to endorse this point 1000%. Brucee, in particular, seems convinced that P2 (or P1.5) will never thrive unless it is as close as possible in price and features to an ARM Cortex M4. I get depressed every time I read such posts. It is a message of utter futility. According to this doctrine, Parallax may as well close up shop.
The Propeller is special, and will always fill a unique role for those who 'get' it. I'm delighted by the new chip even before it is designed. This is entirely because Chip is the one designing it.
BTW, the main problem with the word 'cores' is that certain competing products have watered down the term, as Heater has often explained. Whatever advertising Parallax uses, they need to make certain that the concept is clearly communicated that the chip has 16 physical CPUs, with both individual memory and shared memory. Those are simple concepts that should have broad appeal without a lot of explanation.
Yes, more, faster, better.
Your story is so true. As I have often mentioned I skipped over the Prop many times when surfing through the ELFA catalogue checking out their micro selection . There on the page was that funky hub diagram. I could not relate to any of that or see why I might need one. Eight processors, assuming that is what a "COG" is, sounds interesting. But what? 512 instruction execution space, obviously useless. Next.
User Name,
Yep. I'm really annoyed by a certain competitor bending terms the way they have. On the other hand we have Adapteva famously reinforcing the use of "core" for their multi-processor devices.