VoIP is not optimizable?

When vendors claiming performance improvements, they often refer to a moment in time in history when an improvement was observed on running a specific aplication, a particular transaction on a particular link with an unique environment of latency and packet-drop rate. These are all variables with time. I have observed a 400X (no, not 400%) improvement over the internet doing FTP upload to a public FTP site in Taiwan around 10 am in July with single-ended soft WAN Op client on a XP PC. You just have to be that specific. On another day or simply the next hour, results can vary significantly. On enterprise networks however, you get more consistent results. In reference to Jim's comments about "production network" the public network, the Internet is the production network, for many SMBs.

Stay with Bluecoat if you have a lot of enterprise WEB apps and if you need QoS a lot.Bluecoat and Packeteer are the strongest respectively on these two functions.

Riverbed's WAFS and CIFS are very strong so use these boxes on your backup links to squeez more juice out of the pipes. Delta compression is the key here. I think you have the best combination in the shop here. Don't change it and keep both vendor's reps entertaining you so your company benefits the most in the situation and hence better service.

By the way, RVBD is down to market cap $1.12B just now, and BCSI at $553M. If M&A is talked about, I see latter has higher chance.

I think not.

Look at how many players are there now in the WAN OP market space. With the Gartner forecasted $900M total WW revenue by 2009, how big can this be? Why most WAN Op stocks keep sliding?

My take on the next big thing should be OUTSIDE of the enterprise arena and into the SMB, SOHO, consumer, public Internet, mobile WiFi, WiMAX, and 3G Optimization. Put this in perspective, how are you going to accelerate after you get hom and start going onto the Internet from home on your ADSL? How are you going to go faster on your iPhone 3G? How can you download pictures at 12Mbps inside a car running at 80m/h?

The intention and will to continue investment is one thing, synergy and outcome are another.

Let's read some history right inside the relevent space here. After Workfire acquisition by PKTR, where is the product today? How much revenue contribution by Mentat and Tacit to PKTR revenue after their acquisitions? When an M&A occures it is not only the will of management that counts, but technologies integration, OS compatibility, culture differences, hardware platforms, customer and channel loyalties all fall into play. The ultimate result is not the sum of these variables but the product of them all, in my opinion, and by observations.

WanOp has always been emerging, it just never gets there. Anyone remember the "Time Machine". Build an industry around band-aids for bad protocols? How exactly do you prove these devices aren't corrupting your data in transit?

The industry needs to demand that vendors stop foisting off solutions that use LAN protocols over a WAN. CIFS (really SMB) over a WAN - the definition of a stupid idea. Have you ever send a packet level trace of this protocol? It takes a gazillion end-to-end transfers just to find out what's in a directory. Talk about adding complexity to make something fundamentally ill suited for the purpose work (sort of).