Niel Ransom, Alcatel CTO

Niel Ransom has spent most of his working life figuring out what's going to happen next in telecom.

Armed with a PhD in Electrical Engineering from the University of Notre Dame in the U.S., he started working in applied research at Bell Labs and then moved to BellSouth Corp. (NYSE: BLS) as senior director of advanced technology systems.

While at BellSouth, Ransom got involved in the Joint Procurement Consortium (JPC) – four RBOCs (regional Bell operating companies) that put their combined weight behind the mass deployment of DSL (digital subscriber line) technology in 1996 by jointly negotiating the lowest possible equipment prices.

As luck would have it, Alcatel SA (NYSE: ALA; Paris: CGEP:PA) had some cost-effective DSL technology ready to roll at the time, and it took a bit of a gamble on offering the JPC a rock-bottom price.

To cut a long story short, the gamble paid off: The JPC's optimistic forecasts of DSL volumes turned out to be a considerable underestimate, and Alcatel has become top dog of a monster marketplace, way ahead of everybody else.

Ransom hopped on board in 1997, initially as general manager of Alcatel's local networks business unit, and then as CTO in March 2002 (see Ransom For New Alcatel CTO).

As it happens, DSL history could be about to repeat itself, with another bunch of RBOCs combining forces to roll out fiber-to-the-premises (FTTP) technology in a big way. The RBOCs were aiming to shortlist vendors last Friday (August 8), and Alcatel is viewed as a likely winner (see Vendors Await FTTP Shortlist).

Ransom, of course, needs to think farther ahead – to the opportunities that might turn into major markets in the 18 months or so that it takes to develop new technologies. He also plays a prominent role in picking acquisition targets, the latest of which was TiMetra Networks.

In this interview, Ransom talks about:

— Interviewed by Peter Heywood, Founding Editor, Light Reading

Light Reading: What's the background to the TiMetra acquisition?

Ransom: The overall background is a refocus of Alcatel as we’ve looked at where we need to be in the future to grow and be successful.

We’re really focused on this idea of billable services: How do you create value that end users will pay for? Because that’s ultimately what’s going to grow our industry. And as part of that, it’s clear that Alcatel needs to get stronger in the IP space. We know that. We’ve been investing internally. We’ve done other acquisitions. But it was still clear, especially in the IP edge, that it was a weak area for Alcatel and one we needed to invest in.

So we had Newbridge as a sort of multiservice switch, and we had a lot of investment in enterprise kinds of solutions, but the carrier edge, especially in service creation – being able to do services at wire speeds, virtual private LAN services, VPNs with SLA guarantees, those sorts of things... It looked like it was going to be an important market.

In general, my preference is always to create technology internally. It’s the cheapest way to do R&D – do it yourself, rather than to let someone else do it and then acquire them. Although I must say, with some of the prices that have been happening in the market these days, that can start to become a bargain.

Light Reading: A bargain? Did you read the story we did on your acquisition of TiMetra where we quoted a VC saying the banker representing TiMetra "could have sold ice to Eskimos"? [See Alcatel & TiMetra Seal the Deal.]

Ransom: Yes, I saw that. Of course we looked at TiMetra very closely, and we think we got a heck of a good deal on this, especially compared to other things we looked at in the past. Given all your wire taps (or however Light Reading gets its information), you must have known that we were looking at a few others as well.

And the TiMetra stuff – it's not just that it was good technology; it's a good team. Basil Alwan [TiMetra CEO] is just going to be a tremendous asset, just in himself. He has just the right chemistry to work well with Alcatel. It’s not one of those deals where they’re sort of serial entrepreneurs – they get the acquisition and say, "Now let's go and do the next one." It’s clear to us that they want to be part of Alcatel.

Light Reading: But Alcatel’s had a few bad experiences with acquiring companies, hasn’t it? One example is Packet Engines. I’m sure you read our interview with Bernard Daines. And there are other examples, aren’t there, where acquisitions haven’t worked out for one reason or another?

Ransom: Well that’s right. You have to put a whole lot of emphasis on the chemistry of the people; and that’s why, in the case of TiMetra, it was not sufficient to say: "Is this good technology?" We also had to see whether the customers wanted to buy this technology. We spent a lot of time talking to customers that had the [TiMetra] equipment in their labs – like Masergy – and the feedback we were getting was that they were very, very impressed by the technology.

I should also point out that we’ve had some great examples [of successful acquisitions]. Genesys [acquired in 1999 for $1.5 billion in stock] is a fine example. They’re No. 1 in call centers. We brought them into Alcatel and they’ve thrived. We’ve allowed them to keep their identity.

Light Reading:: Why’s that? Because you’ve changed the way you handle acquisitions?

Ransom: Two things: First, knowing more of what to look for; and secondly, knowing how to handle the acquisitions afterwards.

Light Reading: Right. That seemed to be the problem with the Bernard Daines one, didn’t it? It was all sweetness and light until the Alcatel guys moved in.

Ransom: No. No. We bought a group of startups and brought them together and tried to form a single team of the startups; so again, we were treating them with a degree of hands-off, but we had clustered them together. Maybe something that’s true is that the leaders of startups tend to have unique personalities.

If you take a couple of startups and just reverse the leaders, it’s not clear whether things are going to work.

Light Reading: It's coming back to me now. It was the guy from Xylan that Bernard really couldn't stand, wasn't it? [Steve Kim, referred to as "almost a mad-man" by Daines in his Light Reading interview.]

Ransom: Yes, well... [laughing] The lesson is to preserve the team's identity. It’s not a matter of Big Alcatel versus the little company.

Light Reading: What is it about TiMetra’s technology that marks them out as different from... well, let’s start with the other edge router vendors, with Laurel, for instance. I guess you also looked at Laurel, because it seemed to go through a phase of being up for sale and then not.

Ransom: Well, everything’s for sale. There are, of course, times when companies are desperate to sell. But on TiMetra’s technology, it’s a combination of great density and deep packet inspection and wire rates.

Light Reading: And that’s all to do with being able to do quality-of-service at a very granular level?

Ransom: And the kind of billing and rating that you want to be able to do at the same time. I believe the future is all about creating billable services.

Light Reading: I guess you’ve looked at, or probably crawled all over, Caspian Networks, haven’t you? They also make a big thing out of deep packet inspection, don't they?

Ransom: Yeah. There’s a number of companies with great technology, but with TiMetra I liked not just the technology but also the early market traction [notably with Masergy Communications; see TiMetra Shoots for Service Edge]. Kenny Frank [SVP of engineering and systems at Masergy] used to work for me at BellSouth, and so I’d known him for quite some time.

Light Reading: Oh I see. So that helps a lot. You know somebody well at a company trialing some technology from a company you're thinking of acquiring. You trust them, and...

Ransom: There’s a number of other [TiMetra customers] that haven’t made announcements that I’ve also talked to. It’s a small world. I know the people there. I can call them and say: "Is this good technology? Is this something that might scratch an itch that you have? Is the timing right?"

What we would typically get was: "Wonderful technology. We like the people. Basil is a great guy." We needed something in this space. We needed something with the right kind of technology that was starting to get traction with customers, with a team that we believe in.

Light Reading: Do you see TiMetra as the foundation for some B-RAS [broadband remote access server] developments?

Ransom: Yeah, we’ve looked at that... potentially so. But I also know that we don’t want to sidetrack the team. It’s so easy to take a little team and think there’s all kinds of things you can do with this technology and pull them off in five different directions. So, yes, absolutely you could do a great B-RAS with the product, but that’s not what they’re doing right now. Is that where I’d put my early emphasis on them? Maybe not.

The B-RAS has been a box that sits somewhere deep in the network, but it becomes a bottleneck. So there’s a necessity to distribute the function – in a whole bunch of little B-RASs. [For more on this topic, see Carriers Want a Little B-RAS on the Edge.]

Light Reading: You have a project to integrate B-RAS functionality into your DSLAM, using technology from Corona [Networks Inc.], don’t you? [See Alcatel Unveils 'Better' DSLAM and Corona Gets a Boost.]

Ransom: Yes, and that’s the point I was going to make. You no longer have a centralized B-RAS. You put it, for instance, in the edge routers, and that’s a possibility; or you distribute it all the way down to the DSLAM. And, in fact, these DSLAMs are getting to be pretty big boxes as it is. The 7301 will support 10,000 users. So, we’re saying, gee – maybe we want to do more value creation right there in the DSLAM itself.

Light Reading: So you need to bet on both horses, basically – integrating B-RAS functions in edge routers and DSLAMs?

Ransom: I think there’s going to be a market for both. Or you could say it’s a continuous migration from the core, to starting to be in the edge, to being in the DSLAM itself. And by the DSLAM, I don’t necessarily mean the box with the modems in it, because one of the features of big DSLAMs like the 7301 is that it’s able to host smaller DSLAMs off of it.

With VDSL, it can be nice to have smaller DSLAMs nearer the customer and then aggregrate them with a big DSLAM like the 7301. Would it be nice to have B-RAS functions in that bigger box? I think so.

Light Reading: I guess you’re looking at your dominance in the DSL market and saying, "How can we leverage this?" Do you have all the products to do this?

Ransom: That’s a pretty good question. I may have given some thought to that [laughing].

Alcatel has a lot of strengths, not just in access. We’re No.1 in call centers. We’re No.1 in prepaid billing systems. In each case, we’re saying: What are our strengths? How can we grow using them?

The answer is both vertically and horizontally.

Can I grow vertically? Can I do value creation in the access network so it's not just a dumb DSLAM? Can I start to do things like IP multicast for video?

Can I grow horizontally? Alcatel is dominant in DSL, in digital carrier loop, and is a number one satellite supplier in Europe. Are there other access things we can do?

Light Reading: Are you strong in the cable market, or is that an area you can’t get into because it’s your customers’ competitors?

Ransom: That’s an interesting issue. If you sell arms to the enemy, does that hurt your sale to your current customer or not? I don’t know necessarily that it does. Perhaps it gives a little cachet to the company.

It’s really a question of focus. Where do we spend our energies? Right now, I’d say that the CMTS [cable modem termination system] market is a closed market. The nature of our industry is such that, once an operator chooses a supplier, builds operations systems, trains people... Once they’ve done that, it's very hard for someone else to come in. It works greatly to our advantage sometimes, but when we’re on the outside it’s very hard.

If there’s opportunities... For instance, it’s clear the cable industry is looking a lot now at voice over IP – that might be an opportunity for Alcatel, in that there’s no established softswitch supplier. I’m not saying we would go after that market, but we could. It’s an example of what I mean.

Light Reading: You’re the guy that’s got to place Alcatel's bets on what’s going to be hot in a few years time, aren't you?

Ransom: You know, to some degree predicting the future is easy. It’s predicting when something is going to happen that’s the hard part. If you get to the market too soon, then when the market does open, you have a problem. Your product is architected around technology that was a year older, say, than your competitors'.

Light Reading: It sounds as though there’s a fair percentage of luck in it.

Ransom: That’s right. As they say, "I’d rather be lucky than smart."

Light Reading: Could you give examples where Alcatel got it right in terms of technology and timing, and where it got it wrong?

Ransom: DSL is obviously one we got right, and it’s one where... Gee, do I have to say we were really smart? Or were we lucky?

In that case, if you really look at the history of DSL, BT had decided that it would do video-on-demand using a DSL-like approach – and it had invited Alcatel to put together a network for them. We had built a product for them, which of course went nowhere, because that market never really took off.

However, then this whole JPC came along wanting the same sort of DSL technology – not for video, but for high-speed Internet access. So Alcatel, by luck, had a good, cost-effective product about ready to deploy when other companies were waking up and saying, "Gee, I dunno – what is DSL?"

Light Reading: Gee, I dunno – what is the JPC?

Ransom: It’s the Joint Procurement Consortium. It was four of the RBOCs getting together and saying, if we’re going to make DSL happen, it’s got to be priced right.

A lot of the early DSL products used in trials were $2,000 to $3,000 a line. And we said, if we’re going to deploy this, we need the price significantly down to do that. We’re going to have to drive some volume. Wouldn’t it be better if we got together and bought as one? And that’s what the JPC was all about. [Ransom worked for BellSouth, a founding member of the JPC, along with Ameritech, PacificBell, and SBC Communications.]

In the JPC, we put down certain volume targets for the suppliers, to cause them to salivate and take some risk. It’s the nature of RFPs to always ask for pricing for different volumes, and to have a really big number for the biggest volume, hoping suppliers will put their very best price on that one.

[Carriers] always come back and say: "Well, that’s the price we want, but it’s not our nature to sign up for any guaranteed volumes." They will always put up this big number that you know they have no hope of ever achieving. Well, as it turned out, the big number in the JPC is turning out to be an underestimate of what the market eventually became.

Light Reading: I’ve heard people say that Alcatel has achieved its dominance by being very competitive on price. Was that a decision made early on? As in: "We need to dominate this market – we’ll do it on price."

Ransom: Price was part of the equation. But the other thing that Alcatel had was a working product, because they’d built it for BT to do video.

Light Reading: It’s classic that BT has ended up behind everybody else [on DSL deployment].

Ransom: I wouldn’t say behind. By the way, BT’s research lab [BTexact Technologies] was really good at seeing these trends.

An interesting one right now is PON [passive optical networking]. I would say that BT was really a big pusher of PONs in the past, not that they’ve deployed much. And certainly the PON market is now taking off.

There’s a new JPC of sorts – a common specification between Verizon, BellSouth, and SBC for a fiber-to-the-user-based solution. And, of course, Alcatel is very interested, as others are, in winning that market. [See Vendors Await FTTP Shortlist.]

Light Reading: Do you see the history of the DSL market repeating itself with PONs?

Ransom: Could well be, especially new-build areas. But the interesting thing is that a lot of the early demand has been from the municipalities. A number of them around the world are now allowed to deploy fiber in competition with the carriers. And they say, "We will wait forever if we wait for carriers to deploy fiber in our neighborhood – let’s do it ourselves." That’s becoming the early market demand, and the big carriers are quickly following suit.

Light Reading: With DSL, you lucked out in a sort of way, or grabbed your opportunity. Can you give me an example of Alcatel thinking something big was happening and it not happening – or the timing being wrong?

Ransom: One of the things we thought was going to happen for sure – just looking at what was happening in optics – was that eventually the [electronic] switches would become photonic switches.

We put together not only a photonics crossconnect system, which a number of people did with MEMS [micro-electro-mechanical system] and so forth, but we also had a photonic packet switch in our research lab, using burst switch technology. And it seemed, yes, this is definitely the next thing; and again, it was one of those things where we were absolutely right. This will all happen in the future, but the timing was such that it’s turned out that it’s not going to happen as soon as we thought.

Light Reading: So is there any way of preserving the technology you developed?

Ransom: Of course we didn’t erase all the files. In fact, we keep a level of action on 40-gig transmission systems and some of the crossconnects. But I need to be careful, because when it does happen, there will be other sets of technologies available, and I don’t want our people to say: "Oh, I know how to do that because we built one four years ago, and we’ll just dust that off."

Light Reading: Aren’t the guys that invented the technology for your optical switch developments now working for Avanex [which acquired Alcatel's optical components business], or maybe they've been made redundant?

Ransom: Well, not in photonic switching. We’re getting out of components, but we’re very much into systems. There will be some of the people working on devices that will now be part of Avanex.

You can guess too soon, or you can guess the other way... So this business of deep packet inspection and wirespeed [which Timetra brings to the table], that might be one that we guessed the other way. Suddenly, there it is in front of us. Oh my Goodness! What are we going to do? And in that case, the usual approach has to be: an acquisition. Otherwise you’re going to miss the market.

Light Reading: Are there some other areas where you’ve saying "Oh my Goodness!" and you’re looking to make an acquisition?

Ransom: Oh yeah. I keep reading Light Reading to see what I’m going to do next.

Light Reading: Actually, we've got a lot of Alcatel acquisition predictions right, some of them as much as a year in advance. [See Alcatel Seeks to Buy Optical Startup dated November 8, 2000, and Alcatel Acquires Astral Point -- Cheap dated January 22, 2002.]

Ransom: That’s right [laughing]. If Light Reading says Alcatel is thinking about something, then we say: "Well, if we’re not, maybe we should be."

Light Reading: We were tipped off about the TiMetra acquisition in an anonymous email. Maybe you sent it?

Ransom: [Laughing] I don’t think it was me.

Light Reading: The same person that sent us the note about TiMetra – which proved to be absolutely on target – also sent us a note saying Nortel is sniffing around Procket, by the way [See Will Nortel Pick Procket? .]

Ransom: Really? Well, Procket’s an interesting company.

Light Reading: What's the next BIG thing?

Ransom: Fixed/mobile convergence. All kinds of implications are in that, but we'll have to talk about that some other time.

Light Reading: Ah yes, the timing thing again. The clock has beaten me.