Craig, what I meant is that I see one big difference between Juniper and Cisco.

What I see with the  CISCO messaging is that a vision comes first , few examples:

Cloud computing: not "discovered" by CISCO, is a trend. The real pioneer here is to my eyes GOOGLE but the Unified Computing System is an innovation here b/c it's changing the paradigm in the DC. The message behind is : to go "cloudy" we are not missing processing power or memory or cooling or whatever, we are missing real new connecting networks between systems and hence, we are missing not  systems as such but systems  that can be connected efficiently. Cisco markets UCS which is quite a step ahead. Now, it can be a bad product, it might not function well , it might be a fake I don't know but there's a vision behind it.

Videosurveillance: a sector where cisco is making alliances with Pelco...I don't want to go long here but again, the message is we believe disparate systems, as we have done in the past with telephony will converge on the network.  It might be a naive vision , even one that won't have a future but here again I see a vision.

Now, with Juniper I see the other way around, starting from (good or fantastic, maybe) products to see if putting them together we get value which is bigger than the sum of the parts. At the moment, I didn't see it.

So , again, I'm not questioning juniper's products or spreading FUD around them: maybe they're great but what I see is Juniper refines products making them better if possible...but not visions.

Juniper might have some Cisco envy going, but not as much as you're implying. These chips can take 2 years to develop.  There's no way Trio and MX 3D are just a response to QFP -- or if they are, then Juniper must be the most amazing chip-design shop in history.

(Besides, come to think of it - there's a version of "QFP" that's an off-the-shelf EZchip processor.  Are you counting that as innovation?)

Cloud strategy might be a more apt example, but then again, everybody's talking about cloud.  I think all the vendors are getting pushed into that space.

I think Juniper's bigger problem is that analysts & press think they *should* have Cisco envy.  Does Juniper really need to match Cisco in order to sell routers?  Maybe so, but it's possible we overstate that case sometimes.

Not sure I agree with you there... I think the ability to add features on top of Junos could be a pretty big deal.  It's going to be monitoring & analyzing features at first (like the applications that come with Junos Space) and could get into carrier-tailored versions of those functions, and others, later.

It's not the interaction of devices that's interesting here, it's this ability to 'program' the network.  Who knows what people will do with that.  It's worth watching, I think.

The funny thing about juniper is that they go over everybody's else innovation to my eyes. They might have nice products but almost never a new idea. Basically  psicologically and not only on the competitive market, I think they have a freudian problem with CISCO. Cisco puts the DC at the center and pursue cloud strategy? Juniper make announces they believe in cloud too. Cisco sells new edge routers with ASR 1k and 9K and QFP instead of off-the-shelf asic ? Juniper revamps their edge routers too and guess what, with an purposedly developed ASIC.

Cisco is increasigly investing in blade switches and now in computing? Juniper puts money in Blade Switch. Now, I can even buy  they have nice and cool product less buggy than IOS (but that is not proven) but dear Juniper, when will you make networking make a (even very very small) step ahed with a little bit of innovation?

P.S. If someone thinks an ex Microsoft Exec is  best suit  to sell product that 90% of time don't deliver but present them to the customer as vital upgrades please raise your hand.

i did not find juniper credibly explaining how does an open junos actually help major enterprise and service provider customers??  ip platforms/protocols are open by nature that one can mix various ip devices from various vendors in the network and these devices would interact over common and well defined and open ip protocols.  what more valuable stuff and differentiation do you get when all these ip devices in a large network are running junos??  is that differentiation good enough for customers to make decisions based on it. i doubt it.  perhaps the new ceo from microsoft knows how to market sw credibly & nicely? i doubt that too now....