This time Edward Schneider claims to have made a more thorough analysis, having gone to an even higher-level expert (alas, without naming him), and citing Scott Elder's response to the recent EE Times article.

Schneider mentions a couple of positive things about POET. This is quite clever, because to the unaware reader this suggests a balanced approach. However, his approach isn't balanced at all, and not surprisingly he reaches a negative conclusion again.

The article makes a lot of negative, plain wrong statements about POET, e.g. the brittleness of GaAs wafers. This has been discussed before and is no longer an issue once the chip is sealed in its casing. There are more false statements in the article, but I'll leave them as an exercise to the reader.

On the one hand Schneider claims GaAs chips will be "very expensive to produce compared to those of Si" – a statement which would come closer to the truth if he had replaced "very" by "more", if he would have said that the raw material difference is $1 or so, or if he would have mentioned that the POET process operates with lower temperatures than silicon and is actually shorter in certain respects. On the other hand Schneider has not a single word about the tremendous manufacturing cost savings which are possible by integrating multiple traditional chips into a single POET chip.

Schneider is unbalanced in that he does share the views of his un-named (and possibly non-existing) expert and in that he cites Scott Elder's comment on the EE Times article. However, he omits POET Technologies' response to Elder's comment. A fair, balanced article would have quoted both.

While the longs consider POET's real strength not only in higher speed and lower power consumption, but also in the monolithic integration of optics and electronics, exactly this is what upsets Schneider most. "Unrealistic" he calls POET's business plan relying on GaAs optoelectronic IC (OEIC) logic.

Overall, Schneider and/or his alleged expert cling to the traditional view on GaAs and equalize POET to GaAs, restricting both to niche markets like RF (radio frequency), PA (power amplifier), and LED (light-emitting diode). On the other hand Schneider seems to see the new possibilities somewhat, but talks them small: "POET's dual p-type/n-type transistor advantage over NMOS, leads to lower power consumption versus GaAs peers. Management must focus on the areas where they can leverage this edge." Schneider compares POET to "GaAs peers" only. He fails to see the big picture.

Consequently, he sees potential for POET only in GaAs niche markets, if at all. For POET-based CMOS aka p-type/n-type transistors, he suggest "space stations and satellites". Oh, well!

Bottom line: An unbalanced article with a balanced semblance. Cleverly written, with a negative attitude. Deafening, revealing omissions! Still perfectly able to cast FUD (fear, uncertainty, and doubt) on investors who don't know better.