As someone who has been a customer once and a near-customer another time (both times on fixed line and wireless network tech), Nokia has fantastic, properly bri

As someone who has been a customer once and a near-customer another time (both times on fixed line and wireless network tech), Nokia has fantastic, properly brilliant hardware, and middling software (though moving in the right direction - they're moving to open interfaces, NETCONF on hardware rather than just NBIs on big opaque NMSes they charge a fortune for, etc), and an utterly useless and highly expensive services division they could do well to lose. Big, national telcos don't need "services" divisions like that and the few I know of who are tied into Nokia are desperately trying to get away from the services arm and that usually involves losing the software/hardware too. They're so heavily tied into the sales pitch that it's nigh-impossible to just buy a box now. If the services arm were competent they'd have a chance but stuff like their network design and project management services are just flat-out incompetent. Compare that to a lot of smaller players (e.g. Edge-Core, IP Infusion) who will ship hardware with open interfaces, may provide NMSes and element management systems etc but generally with open interfaces, and provide as much support as you want to pay for, and for most small and large telcos that's the way to go towards. If I were running Nokia right now I'd be ditching the bulk of the services arm and taking what's good in there into support and engineering, pivoting as quickly as possible to adopt open standards and interfaces, and getting involved with things like the Telecoms Infrastructure Project, CORD, et. al. Their hardware is fantastic but their software sabotages it. This isn't an uncommon pattern in telecoms per se; one might go so far as to call it the default position, especially these days where software is the key part of most fixed/cabled networks, given photonics is broadly standardised below and even at 400G outside of long-haul coherent stuff. Nokia however seem to manage to suffer much more from this than other vendors of kit. Compare with companies like Adtran who have okay hardware, okay software, and are just dominating the entire EU XGS-PON market despite Nokia's desperate attempts to claim it; mostly this is down to TCO for the Nokia stack being a bit higher, and the whole services/support/software arm being so huge and unwieldy. You can buy a $20k line terminal but need $50k of software just to turn the damn thing on.